EP2734643A2 - Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression - Google Patents

Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression

Info

Publication number
EP2734643A2
EP2734643A2 EP12751371.1A EP12751371A EP2734643A2 EP 2734643 A2 EP2734643 A2 EP 2734643A2 EP 12751371 A EP12751371 A EP 12751371A EP 2734643 A2 EP2734643 A2 EP 2734643A2
Authority
EP
European Patent Office
Prior art keywords
pathway
tissue
medical subject
activity
target genes
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
EP12751371.1A
Other languages
German (de)
French (fr)
Other versions
EP2734643B1 (en
Inventor
Wilhelmus Franciscus Johannes Verhaegh
Anja Van De Stolpe
Hendrik Jan VAN OOIJEN
Kalyana Chakravarthi DULLA
Marcia Alves De Inda
Ralf Hoffmann
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Koninklijke Philips NV
Original Assignee
Koninklijke Philips NV
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Koninklijke Philips NV filed Critical Koninklijke Philips NV
Priority to EP20209212.8A priority Critical patent/EP3831957B1/en
Priority to EP12751371.1A priority patent/EP2734643B1/en
Priority to EP18180343.8A priority patent/EP3418396B1/en
Priority to DK16192584.7T priority patent/DK3173489T3/en
Priority to EP22177533.1A priority patent/EP4130291B1/en
Priority to EP16192584.7A priority patent/EP3173489B1/en
Publication of EP2734643A2 publication Critical patent/EP2734643A2/en
Application granted granted Critical
Publication of EP2734643B1 publication Critical patent/EP2734643B1/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6809—Methods for determination or identification of nucleic acids involving differential detection
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6876—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes
    • C12Q1/6883—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material
    • C12Q1/6886—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material for cancer
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models
    • G06N5/048—Fuzzy inferencing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
    • G16B20/20—Allele or variant detection, e.g. single nucleotide polymorphism [SNP] detection
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
    • G16B20/30—Detection of binding sites or motifs
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B25/00—ICT specially adapted for hybridisation; ICT specially adapted for gene or protein expression
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B40/00—ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B40/00—ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
    • G16B40/20—Supervised data analysis
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B5/00—ICT specially adapted for modelling or simulations in systems biology, e.g. gene-regulatory networks, protein interaction networks or metabolic networks
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B5/00—ICT specially adapted for modelling or simulations in systems biology, e.g. gene-regulatory networks, protein interaction networks or metabolic networks
    • G16B5/20—Probabilistic models
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/106—Pharmacogenomics, i.e. genetic variability in individual responses to drugs and drug metabolism
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/112—Disease subtyping, staging or classification
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/158—Expression markers

Definitions

  • the subject matter described herein mainly relates to bio informatics, genomic processing arts, proteomic processing arts, and related arts.
  • Genomic and proteomic analyses have substantial realized and potential promise for clinical application in medical fields such as oncology, where various cancers are known to be associated with specific combinations of genomic mutations/ variations and/or high or low expression levels for specific genes, which play a role in growth and evolution of cancer, e.g. cell proliferation and metastasis.
  • the Wnt signaling pathway affects regulation of cell proliferation, and is highly regulated.
  • High Wnt pathway activity due to loss of regulation has been correlated to cancer, among which with malignant colon tumors. While not being limited to any particular theory of operation, it is believed that deregulation of the Wnt pathway in malignant colon cells leads to high Wnt pathway activity that in turn causes cell proliferation of the malignant colon cells, i.e. spread of colon cancer.
  • abnormally low pathway activity might also be of interest, for example in the case of osteoporosis.
  • One approach is to limit the analysis to a few canonical or standardized tests, such as tests approved by the U.S. Food and Drug Administration (FDA).
  • FDA U.S. Food and Drug Administration
  • a specific indicator or combination of indicators e.g., mutations and/or specified high or low gene expression levels
  • the canonical test is supported by clinical studies that have shown strong correlation with the disease condition or with treatment efficacy. This approach is useful only for those clinical conditions for which a canonical test has been developed, e.g. specific diagnosis of a disease, or predicting response to a drug in a specific cancer type at a specific stage, and is also rigid as it is only applicable for the canonical conditions.
  • the Wnt pathway comprises a cascade of proteomic reactions.
  • Major components of this chain include (but are not limited to) binding of the Wnt signaling protein to a frizzled surface receptor of the cell which causes activation of proteins of the disheveled family of proteins which in turn impact the level of transcription agents such as -catenin/TCF4 based protein complexes in the cell nucleus.
  • downstream may be inapplicable for a substantial portion of the Wnt pathway; more generally, abnormal behavior in one portion of the protein cascade comprising the Wnt pathway may have more or less effect on other portions of the protein cascade, and on the activity of the Wnt pathway as a whole. Still further, in some clinical studies protein expression levels for regulatory proteins of the signaling cascade are assessed by measuring mR A expression levels of the genes that encode for the regulatory proteins. This is an indirect measurement that may not accurately assess the regulatory protein expression level, and hardly ever reflects the amount of active proteins (after a specific post-translational modification like phosphorylation).
  • the main problem underlying the present invention was thus to provide suitable methods and means for performing genomic and, respectively, proteomic analyses Specific aspects of the underlying problem as well as further objections in connection with the present invention become apparent when studying the description, the examples provided herein and, in particular, when studying the attached claims.
  • the present invention provides new and improved methods and apparatuses as disclosed herein.
  • the above problem is solved by a specific method for assessing cellular signaling pathway activity using probabilistic modeling of target gene expression, namely a method comprising:
  • inferring activity of one or more cellular signaling pathway(s) in tissue of a medical subject based at least on the expression level(s) (in particular on mRNA and/or protein level) of one or more target gene(s) of the cellular signaling pathway(s) measured in an extracted sample of the tissue of the medical subject, wherein the inferring comprises inferring activity of the cellular signaling pathway(s) in the tissue of the medical subject by evaluating at least a portion of a probabilistic model, preferably a Bayesian network, representing the cellular signaling pathway(s) for a set of inputs including at least the expression level(s) of the one or more target genes of the cellular signaling pathway(s) measured in the extracted sample of the tissue of the medical subject;
  • a probabilistic model preferably a Bayesian network
  • TF transcription factor
  • the inferring is performed by a digital processing device using the probabilistic model of the cellular signaling pathway(s).
  • target gene(s) may be "direct target genes” and/or “indirect target genes” (as described herein).
  • the inferring comprises estimating a level in the tissue of the medical subject of at least one transcription factor (TF) element represented by a TF node of the probabilistic model, the TF element controlling transcription of the one or more target gene(s) of the cellular signaling pathway(s), the estimating being based at least in part on conditional probabilities of the probabilistic model relating the TF node and nodes in the probabilistic model representing the one or more target gene(s) of the cellular signaling pathway(s) measured in the extracted sample of the tissue of the medical subject.
  • TF transcription factor
  • the probabilistic model may be a Bayesian network model.
  • the inferring is performed by using a Bayesian network comprising nodes representing information about the signaling pathway(s) and conditional probability relationships between connected nodes of the Bayesian network.
  • the cellular signaling pathway(s) may be a Wnt pathway, an ER (Estrogen Receptor) pathway, an AR (Androgen Receptor) pathway and/or a Hedgehog pathway.
  • the cellular signaling pathway(s) comprise(s) a Wnt pathway, an ER pathway, an AR pathway and/or a Hedgehog pathway.
  • the target gene(s) is/are selected from the group comprising or consisting of target genes listed in Table 1 or Table 6 (for Wnt pathway), target genes listed in Table 2, Table 5 or Table 7 (for ER pathway), target genes listed in Table 3 or Table 8 (for Hedgehog pathway) and target genes listed in Table 4 or Table 9 (for AR pathway).
  • the inferring comprises:
  • inferring activity of a Wnt pathway in the tissue of the medical subject based at least on expression levels of one or more, preferably at least three, target gene(s) of the Wnt pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: KIAA1199, AXIN2, RNF43, TBX3, TDGF1, SOX9, ASCL2, IL8, SP5, ZNRF3, KLF6, CCND1, DEFA6 and FZD7.
  • the inferring is further based on expression levels of at least one target gene of the Wnt pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: NKD1, OAT, FAT1, LEF1, GLUL, REG1B, TCF7L2, COL18A1, BMP7, SLC1A2, ADRA2C, PPARG, DK 1, HNF1A and LECT2.
  • the inferring also comprises: inferring activity of an ER pathway in the tissue of the medical subject based at least on expression levels of one or more, preferably at least three, target gene(s) of the ER pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: CDH26, SGK3, PGR, GREBl, CA12, XBP1, CELSR2, WISP2, DSCAM, ERBB2, CTSD, TFF1 and NRIPl .
  • the inferring is further based on expression levels of at least one target gene of the ER pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: AP1B1, ATP5J, COL18A1, COX7A2L, EBAG9, ESR1, HSPB1, IGFBP4, KRT19, MYC, NDUFV3, PISD, PREDM15, PTMA, RARA, SOD1 and TRIM25.
  • inferring activity of a Hedgehog pathway in the tissue of the medical subject based at least on expression levels of one or more, preferably at least three, target gene(s) of the Hedgehog pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: GLI1, PTCH1, PTCH2, IGFBP6, SPP1, CCND2, FST, FOXL1, CFLAR, TSC22D1, RAB34, S100A9, S100A7, MYCN, FOXM1, GLI3, TCEA2, FYN and CTSL1,
  • the inferring is further based on expression levels of at least one target gene of the Hedgehog pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: BCL2, FOXA2, FOXF1, G19, HHIP, IL1R2, JAG2, JUP, MIF, MYLK, NKX2.2, NKX2.8, PITRM1 and TOM1.
  • a method wherein the inferring (also) comprises
  • inferring activity of an AR pathway in the tissue of the medical subject based at least on expression levels of one or more, preferably at least three, target gene(s) of the AR pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: KLK2, PMEPA1, TMPRSS2, NKX3 1, ABCC4, KLK3, FKBP5, ELL2, UGT2B15, DHCR24, PPAP2A, NDRG1, LRIG1, CREB3L4, LCP1, GUCY1A3, AR and EAF2,
  • the inferring is further based on expression levels of at least one target gene of the AR pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: APP, NTS, PLAU, CDKN1A, DRG1, FGF8, IGF1, PRKACB, PTPN1, SGK1 and TACC2.
  • Another aspect of the present invention relates to a method (as described herein), further comprising:
  • the recommending is performed only if the cellular signaling pathway(s) is/are determined to be operating abnormally in the tissue of the medical subject based on the inferred activity of the cellular signaling pathway(s).
  • the present invention also relates to a method (as described herein) comprising:
  • inferring activity of an AR pathway in tissue of a medical subject based at least on expression levels of two, three or more target genes of a set of target genes of the AR pathway measured in an extracted sample of the tissue of the medical subject.
  • the set of target genes of the Wnt pathway includes at least nine, preferably all target genes selected from the group comprising or consisting of: KIAAl 199, AXIN2, RNF43, TBX3, TDGF1, SOX9, ASCL2, IL8, SP5, ZNRF3, KLF6, CCND1, DEFA6 and FZD7,
  • the set of target genes of the ER pathway includes at least nine, preferably all target genes selected from the group comprising or consisting of: CDH26, SGK3, PGR, GREB1, CA12, XBP1, CELSR2, WISP2, DSCAM, ERBB2, CTSD, TFF1 and NRIPl, and/or
  • the set of target genes of the Hedgehog pathway includes at least nine, preferably all target genes selected from the group comprising or consisting of: GLI1, PTCH1, PTCH2, IGFBP6, SPP1, CCND2, FST, FOXL1, CFLAR, TSC22D1, RAB34, S100A9, S100A7, MYCN, FOXM1, GLI3, TCEA2, FYN and CTSL1,
  • the set of target genes of the AR pathway includes at least nine, preferably all target genes selected from the group comprising or consisting of: KLK2, PMEPA1, TMPRSS2, NKX3 1, ABCC4, KLK3, FKBP5, ELL2, UGT2B15, DHCR24, PPAP2A, NDRG1, LRIG1, CREB3L4, LCP1, GUCY1A3, AR and EAF2.
  • the set of target genes of the Wnt pathway further includes at least one target gene selected from the group comprising or consisting of: NKD1, OAT, FAT1, LEF1, GLUL, REGIB, TCF7L2, COL18A1, BMP7, SLC1A2, ADRA2C, PPARG, DKK1, HNF1A and LECT2,
  • the set of target genes of the ER pathway further includes at least one target gene selected from the group comprising or consisting of: AP1B1, ATP5J, COL18A1,
  • COX7A2L COX7A2L, EBAG9, ESR1, HSPB1, IGFBP4, KRT19, MYC, NDUFV3, PISD,
  • the set of target genes of the Hedgehog pathway further includes at least one target gene selected from the group comprising or consisting of: BCL2, FOXA2, FOXFl, G19, HHIP, IL1R2, JAG2, JUP, MIF, MYLK, NKX2.2, NKX2.8, PITRM1 and TOM1, and/or
  • the set of target genes of the AR pathway further includes at least one target gene selected from the group comprising or consisting of: APP, NTS, PLAU, CDK 1 A, DRG1, FGF8, IGF1, PRKACB, PTPN1, SGK1 and TACC2,
  • the sample(s) to be used in accordance with the present invention can be, e.g., a sample obtained from a breast lesion, or from a colon of a medical subject known or suspected of having colon cancer, or from a liver of a medical subject known or suspected of having liver cancer, or so forth, preferably via a biopsy procedure or other sample extraction procedure.
  • the tissue of which a sample is extracted may also be metastatic tissue, e.g. (suspected) malignant tissue originating from the colon, breast, liver, or other organ that has spread outside of the colon, breast, liver, or other organ.
  • the tissue sample may be circulating tumor cells, that is, tumor cells that have entered the bloodstream and may be extracted as the extracted tissue sample using suitable isolation techniques.
  • Another disclosed aspect of the present invention pertain the use of a non- transitory storage medium as described herein or a computer program as described herein for specific diagnosis of a disease or predicting response to a drug in a specific cancer type at a specific stage.
  • an apparatus comprising a digital processor configured to perform a method according to the invention as described herein.
  • a non-transitory storage medium stores instructions that are executable by a digital processing device to perform a method according to the invention as described herein.
  • the non-transitory storage medium may be a computer-readable storage medium, such as a hard drive or other magnetic storage medium, an optical disk or other optical storage medium, a random access memory (RAM), read-only memory (ROM), flash memory, or other electronic storage medium, a network server, or so forth.
  • the digital processing device may be a handheld device (e.g., a personal data assistant or smartphone), a notebook computer, a desktop computer, a tablet computer or device, a remote network server, or so forth.
  • a computer program comprises program code means for causing a digital processing device to perform a method according to the invention as described herein.
  • the digital processing device may be a handheld device (e.g., a personal data assistant or smartphone), a notebook computer, a desktop computer, a tablet computer or device, a remote network server, or so forth.
  • CDS clinical decision support
  • Another advantage resides in improved assessment of cellular signaling pathway activity that is less susceptible to error.
  • Another advantage resides in providing a CDS system recommending targeted treatment for loss of regulation of a cellular signaling pathway.
  • Another advantage resides in providing a CDS system that is designed to detect loss of regulation for a particular cellular signaling pathway, such as a Wnt pathway, an ER pathway, an AR pathway or a Hedgehog pathway, and is readily adapted to provide recommendations for different types of cancer sourced by that particular cellular signaling pathway.
  • a particular cellular signaling pathway such as a Wnt pathway, an ER pathway, an AR pathway or a Hedgehog pathway
  • the present invention as described herein can, e.g., also advantageously be used in connection with - diagnosis based on predicted (inferred) activity;
  • Figure 1 shows a simple Bayesian network representing part of a cellular signaling pathway.
  • the cellular signaling pathway is symbolized by a transcription factor (TF) complex and the target genes produced as a result of the transcription factor complex being present.
  • the probabilistic relationship between the TF element and a target gene in case of binary discretization can be represented by a conditional probability table as depicted in the diagram.
  • Figure 2 shows an illustrative Bayesian network describing a hypothetical cellular signaling pathway. Both the upstream proteins and downstream target mR A nodes are depicted in the diagram. The upstream proteins serve as input in the transcription factor complex, whereas the target mRNAs are the transcription factor complex's output nodes.
  • FIG. 1 shows an illustrative example of a Bayesian network representation of a single cellular signaling pathway with multiple transcription factor complex or a multiple cellular signaling pathways with their own transcription factor complex combined into one Bayesian network or a combination thereof.
  • FIG. 1 shows an example of a Bayesian network illustrating a simple representation of a cellular signaling pathway similar to Figure 1. Now additional nodes have been attached to represent the translation of target mRNA into target proteins.
  • FIG. 1 shows the illustrative Bayesian network of Figure 1 with an additional layer of nodes representing the probesets on a microarray chip connecting the probes' intensities to the corresponding target mRNA levels.
  • Figure 9 shows a predicted Wnt pathway activity of the Bayesian network
  • Figure 10 shows a predicted Wnt pathway activity of the Bayesian network and nearest centroid method as described herein in a data set of colon samples (GSE15960).
  • Figure 11 shows a predicted Wnt pathway activity of the Bayesian network and
  • Figure 12 shows a predicted Wnt pathway activity of the Bayesian network
  • Figure 13 shows a predicted Wnt pathway activity of the Bayesian network
  • Figure 14 shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of colon samples (GSE20916).
  • Figure 15 shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of colon samples (GSE4183).
  • Figure 16 shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of colon samples (GSE15960). shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of breast cancer samples (GSE12777). shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of liver cancer samples (GSE9843).
  • FIG. 10327 shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of medulloblastoma samples (GSE 10327).
  • CDS clinical decision support
  • FIG. 12777 shows a predicted ER pathway activity in breast cancer cell lines from GSE12777. shows a predicted ER pathway activity in breast cancer samples from GSE12276. shows a predicted ER pathway activity in cancer cell lines from GSE36133. shows a predicted Hedgehog pathway activity in cancer cell lines from GSE34211. shows a predicted Hedgehog pathway activity in medulloblastoma samples from GSE10327. shows a predicted Hedgehog pathway activity in breast cancer samples from GSE12276. shows a predicted ER pathway activity in MCF7 and Tamoxifen resistant cell lines from GSE21618. shows a predicted ER pathway activity in a time series of estrogen- stimulated MCF7 cell line samples from GSE11324.
  • FIG. 1 shows Wnt, ER and Hedgehog pathway activity in luminal A samples of GSE12276. shows Wnt, ER and Hedgehog pathway activity in basal samples of GSE12276. shows a predicted Wnt pathway activity in colon samples from GSE20916. shows a predicted ER pathway activity in MCF7 cell lines stimulated with estrogen (E2) or a negative control (EtOH) (GSE9253).
  • Figure 37 shows Kaplan-Meier survival curves of patients from the GSE12276 data set grouped according to pathway activity.
  • Figure 38 shows a predicted AR pathway activity in LNCaP cell lines treated with different treatment regimes from GSE7708.
  • Figure 39 shows a predicted AR pathway activity in prostate cancer samples from
  • Figure 40 shows a predicted AR pathway activity in breast cancer samples from
  • Figure 41 shows a predicted AR pathway activity in the GSE36133 data set containing cell line samples representing various cancer types.
  • Figure 42 shows a predicted AR pathway activity in the GSE34211 data set containing cell line samples representing various cancer types.
  • the following examples merely illustrate particularly preferred methods and selected aspects in connection therewith.
  • the teaching provided therein may be used for constructing several tests and/or kits, e.g. to detect, predict and/or diagnose the abnormal activity of one or more cellular signaling pathways.
  • drug prescription can advantageously be guided, drug prediction and monitoring of drug efficacy (and/or adverse effects) can be made, drug resistance can be predicted and monitored, e.g. to select subsequent test(s) to be performed (like a companion diagnostic test).
  • drug resistance can be predicted and monitored, e.g. to select subsequent test(s) to be performed (like a companion diagnostic test).
  • the following examples are not to be construed as limiting the scope of the present invention.
  • Example 1 Bayesian network construction
  • a probabilistic model e.g., the illustrative Bayesian model shown in Figure 6
  • conditional probabilistic relationships between expression levels of a number of different target genes and the activity of the cellular signaling pathway can be used to determine the activity of the cellular signaling pathway with a high degree of accuracy.
  • the probabilistic model can be readily updated to incorporate additional knowledge obtained by later clinical studies, by adjusting the conditional probabilities and/or adding new nodes to the model to represent additional information sources. In this way, the probabilistic model can be updated as appropriate to embody the most recent medical knowledge.
  • the transcription factor complex element is a representation of the level of the transcription factor complex.
  • the protein level of the transcription factor element is connected to a number of mR A levels of the transcription factor's target genes (in this exemplary Bayesian network only three target genes are depicted, that are known to be expressed in the tissue in case the transcription factor is available). It is to be understood that many, most, or all of the pathway's target genes (in case of the Wnt, ER, Hedgehog and AR pathways particularly the target genes mentioned in Table 1, Table 2, Table 3 and Table 4 respectively) are analogously regulated by the TF element.
  • the relationships between the level of the TF element and the mRNA levels of the target genes are modeled in the Bayesian network by the edges. For each of the target genes, a conditional probabilistic distribution specifies how the gene's mRNA level depends on the level of the TF element.
  • the levels of the TF element and target genes may be variously represented.
  • One option is to use a binary discretization, into states “absent” and “present” for the TF element, and “down” and “up” for a target gene's mRNA level (see Figure 1).
  • the probabilistic relationship between the TF element and a target gene can then be represented by a conditional probability table (as indicated in the same figure).
  • levels can also represented as continuous level values, or as quantized values having three or more quantization levels (e.g. "down”, "normal” and “up” for target genes).
  • a Bayesian network model comprises a directed, acyclic graph comprising nodes connected by edges. Each node represents an information item pertaining to the pathway at hand (or, more generally, to the cellular signaling pathway). Pathway element nodes each represent a genomic or proteomic element of the cellular signaling pathway.
  • a pathway element node may represent one of but not restricted to: a protein, a protein complex, an mR A molecule transcribed from a target gene of the cellular signaling pathway, a methylated gene, a phosphorylated protein, a phosphorylated protein complex, or so forth.
  • a protein a protein complex
  • an mR A molecule transcribed from a target gene of the cellular signaling pathway
  • a methylated gene a phosphorylated protein
  • a phosphorylated protein complex a phosphorylated protein complex
  • TF transcription factor
  • the transcription factor (TF) element is believed to be a protein complex (that is, a combination of proteins bound together in a specific structure that performs the function of regulating transcription from the target genes) in the majority of the signaling pathways.
  • the TF element may be a single protein.
  • signaling pathways may exert their activity through more than one transcription factor resulting in a more complex Bayesian network with multiple transcription factors feeding into the target gene(s) (see Figure 3 for a hypothetical illustration of multiple transcription factor elements influencing target gene transcription).
  • Such a multi- transcription factor Bayesian network may also be the result of a combination of pathways combined in one Bayesian network.
  • Additional information nodes further downstream of the target genes may be included in the Bayesian network as well.
  • An illustrative example of this is the translation of target gene's mRNA into proteins ( Figure 4) or protein level nodes of the target gene as surrogate node of the target gene's mRNA level ( Figure 5).
  • the mRNA molecules of the target gene are translated by interaction with ribosome molecules to form proteins corresponding to the mRNA molecules and corresponding to the target genes. This is expression of the target genes at the protein level. Measurement of the protein level by, but not limited to, for example mass-spectrometry, immunohistochemistry, gel electrophoresis techniques may act as evidence for these target protein levels.
  • the expression level of a target gene may be computed based on the measured intensity of corresponding probesets of a microarray, for example by averaging or by other means of other techniques (e.g. R A sequencing). In some embodiments this computation is integrated into the Bayesian network, by extending the Bayesian network with a node for each probeset that is used and including an edge running to each of these "measurement" nodes from the corresponding target gene node, as described herein with reference to Figure 6.
  • the probabilistic model may optionally also incorporate additional genomic information, such as information on mutations, copy number variations, gene expression, methylation, translocation information, or so forth, which change genomic sequences which are related to the signaling cascade of the pathway to infer the pathway activity and to locate the defect in the Wnt pathway which causes the aberrant functioning (either activation or inactivity), as described by illustrative reference to Figure 7 for the illustrative case of methylation and copy number data.
  • genomic information can be available through, but not limited to, RNA sequencing and SNP analysis.
  • the approaches for cellular signaling pathway analysis disclosed herein are readily applied to other cellular signaling pathways besides these pathways, such as to intracellular signaling pathways with receptors in the cell membrane (e.g., the Notch, HER2/PI3K, TGFbeta, EGF, VEGF, and TNF-NFkappaB cellular signaling pathways) and intracellular signaling pathways with receptors inside the cell (e.g., progesterone, retinoic acid, and vitamin D cellular signaling pathways).
  • intracellular signaling pathways with receptors in the cell membrane e.g., the Notch, HER2/PI3K, TGFbeta, EGF, VEGF, and TNF-NFkappaB cellular signaling pathways
  • intracellular signaling pathways with receptors inside the cell e.g., progesterone, retinoic acid, and vitamin D cellular signaling pathways.
  • the performance of two types of machine learning techniques are compared to each other with the Wnt pathway taken as an example case: the prediction of Wnt activity by means of a nearest centroid method is compared to the method of choice according to the present invention, which e.g. uses a Bayesian network.
  • the Bayesian network approach was selected based on its advantages residing in the probabilistic approach being able to incorporate the available information in either "soft”, e.g. percentages of study subjects exhibiting probative characteristics, and "hard” form, using conditional probabilistic relationships.
  • the probabilistic model also enables information to be incorporated based on partial (rather than comprehensive) knowledge of the underlying cellular signaling pathway, again through the use of conditional probability tables.
  • the two centroids are calculated on the same list of probesets used in the Bayesian network, and for the 'Wnt on' and 'Wnt off centroid they are based on the adenoma samples and the normal colon samples, respectively, of the same fRMA processed data of GSE8671.
  • the log2 -ratio of the two Euclidean distances between a sample and the two centroids was subsequently used to classify samples from various data sets to infer the classification of the samples.
  • a log2 -ratio of 0 corresponds to an equal distance of the sample to the two centroids, a value > 0 corresponds to a sample classified as active Wnt signaling whereas a value ⁇ 0 corresponds to a sample identified as having an inactive Wnt signaling pathway.
  • the Bayesian network was constructed similar to Figure 6 and the procedure described herein. Similarly to this description of the Wnt Bayesian network, the conditional probability tables of the edges between probesets and their respective genes were trained using fRMA processed data of 32 normal colon samples and 32 adenoma samples from data set GSE8671 from the Gene Expression Omnibus (accessible at http://www.ncbi.nlm.nih.gov/geo/, last accessed July 13, 2011). The trained Bayesian network was then tested on various data sets to infer the probability P(Wnt On) that the Wnt pathway is "on", that is, active, which is taken equal to the inferred probability that the Wnt pathway transcription complex is "present”.
  • the classification could be corrected by lowering and increasing the threshold of the nearest centroid classification.
  • the idea behind this would be that the threshold of Wnt activity might be altered in different tissue-types. However, this would involve additional training of the nearest centroid method to be applicable to other tissue types.
  • One of the strengths of the Bayesian network model is that this tissue-specific training is not required as it is established to be nonspecific regarding tissue-type.
  • a transcription factor is a protein complex (that is, a combination of proteins bound together in a specific structure) or a protein that is able to regulate transcription from target genes by binding to specific DNA sequences, thereby controlling the transcription of genetic information from DNA to mRNA.
  • the mRNA directly produced due to this action of the transcription complex is herein referred to as a "direct target gene”.
  • Pathway activation may also result in more secondary gene transcription, referred to as "indirect target genes”.
  • Bayesian network models as exemplary probabilistic models
  • direct target genes as direct links between pathway activity and mRNA level
  • Specific pathway mRNA target genes were selected from the scientific literature, by using a ranking system in which scientific evidence for a specific target gene was given a rating, depending on the type of scientific experiments in which the evidence was accumulated. While some experimental evidence is merely suggestive of a gene being a target gene, like for example a mRNA increasing on an microarray of an embryo in which it is known that the Hedgehog pathway is active, other evidence can be very strong, like the combination of an identified pathway transcription factor binding site and retrieval of this site in a chromatin immunoprecipitation (ChIP) assay after stimulation of the specific pathway in the cell and increase in mRNA after specific stimulation of the pathway in a cell line.
  • ChoIP chromatin immunoprecipitation
  • ChIP experiments in which direct binding of a pathway-transcription factor to its binding site on the genome is shown.
  • Example: By using chromatin- immunoprecipitation (ChIP) technology subsequently putative functional TCF4 transcription factor binding sites in the DNA of colon cell lines with and without active Wnt pathway were identified, as a subset of the binding sites recognized purely based on nucleotide sequence. Putative functionality was identified as ChlP- derived evidence that the transcription factor was found to bind to the DNA binding site.
  • Electrophoretic Mobility Shift (EMSA) assays which show in vitro binding of a transcription factor to a fragment of DNA containing the binding sequence.
  • EMSA-based evidence is less strong, since it cannot be translated to the in vivo situation. 3. Stimulation of the pathway and measuring mRNA profiles on a microarray or using RNA sequencing, using pathway-inducible cell lines and measuring mRNA profiles measured several time points after induction - in the presence of cycloheximide, which inhibits translation to protein, thus the induced mRNAs are assumed to be direct target genes.
  • Example for the Wnt pathway Using the known TCF4- beta catenin transcription factor DNA binding sequence, a software program was run on the human genome sequence, and potential binding sites were identified, both in gene promoter regions and in other genomic regions.
  • points can be given incrementally, meaning one technology 1 point, second technology adds a second point, and so on. Using this relatively ranking strategy, one can make a list of most reliable target genes.
  • ranking in another way can be used to identify the target genes that are most likely to be direct target genes, by giving a higher number of points to the technology that provides most evidence for an in vivo direct target gene, in the list above this would mean 8 points for experimental approach 1), 7 to 2), and going down to one point for experimental approach 8.
  • Such a list may be called "general target gene list”.
  • the "general target gene list” probably contains genes that are more tissue specific, and can be potentially used to optimize and increase sensitivity and specificity of the model for application at samples from a specific tissue, like breast cancer samples.
  • Gene promoter/enhancer region contains an estrogen response element (ERE) motif:
  • the ERE motif should be proven to respond to estrogen, e.g., by means of a transient transfection assay in which the specific ERE motif is linked to a reporter gene, and
  • the presence of the ERE motif should be confirmed by, e.g., an enriched motif analysis of the gene promoter/enhancer region.
  • ER (differentially) binds in vivo to the promoter/enhancer region of the gene in question, demonstrated by, e.g., a ChlP/CHIP experiment or a chromatin immunoprecipitation assay:
  • ER is proven to bind to the promoter/enhancer region of the gene when the ER pathway is active
  • b. does not bind (or weakly binds) to the gene promoter/enhancer region of the gene if the ER pathway is not active.
  • the gene is differentially transcribed when the ER pathway is active, demonstrated by, e.g.,
  • RNA Pol II binds to the promoter region of the gene through an immunoprecipitation assay.
  • a suitable experiment for collecting evidence of ER differential binding is to compare the results of, e.g., a ChlP/CHIP experiment in a cancer cell line that responds to estrogen (e.g., the MCF-7 cell line), when exposed or not exposed to estrogen. The same holds for collecting evidence of mRNA transcription.
  • the target genes of the ER pathway used for the Bayesian network model of the ER pathway described herein contain a selection of target genes based on their literature evidence score; only the target genes with the highest evidence scores (preferred target genes according to the invention) were added to this short list.
  • the full list of ER target genes, including also those genes with a lower evidence score, is shown in Table 5.
  • a further subselection or ranking of the target genes of the Wnt, ER, Hedgehog and AR pathways shown in Table 1, Table 2, Table 3 and Table 4 was performed based on a combination of the literature evidence score and the odds ratios calculated using the trained conditional probability tables linking the probeset nodes to the corresponding target gene nodes.
  • the odds ratio is an assessment of the importance of the target gene in inferring activity of the pathways. In general, it is expected that the expression level of a target gene with a higher odds ratio is likely to be more informative as to the overall activity of the pathway as compared with target genes with lower odds ratios.
  • the list of Wnt target genes constructed based on literature evidence following the procedure described herein (Table 1) is compared to another list of target genes not following above mentioned procedure.
  • the alternative list is a compilation of genes indicated by a variety of data from various experimental approaches to be a Wnt target gene published in three public sources by renowned labs, known for their expertise in the area of molecular biology and the Wnt pathway.
  • the alternative list is a combination of the genes mentioned in table S3 from Hatzis et al.
  • Target gene Reference # Target gene Reference #
  • the next step consisted of finding the probesets of the Affymetrix® GeneChip Human Genome U133 Plus 2.0 array that corresponds with the genes. This process was performed using the Bioconductor plugin in R and manual curation for the probesets relevance based on the UCSC genome browser, thereby removing e.g. probesets on opposite strands or outside gene exon regions. For two of the 124 genes there are no probesets available on this microarray-chip and therefore could not be inserted in the Bayesian network, these are LOC283859 and WNT3A. In total 287 probesets were found to correspond to the remaining 122 genes (Table 1 1).
  • the broad literature model generally predicts more extreme probabilities for Wnt signaling being on or off.
  • the alternative model predicts similar results for the colon cancer data sets (GSE20916, GSE4183, GSE15960), but more than expected samples with predicted active Wnt signaling in breast cancer (GSE 12777), liver cancer (GSE9843) and medulloblastoma sample (GSE 10327) data sets.
  • the broad literature target genes list results in approximately equally well predictions of Wnt activity in colon cancer on the one hand, but worse predictions (too many false positives) in other cancer types on the other hand.
  • non-Wnt-specific target genes possibly included in these lists may be a source of the worsened predictions of Wnt activity in other cancer types.
  • the alternative list is likely to contain more indirectly regulated target genes, which probably makes it more tissue specific.
  • the original list is tuned towards containing direct target genes, which are most likely to represent genes that are Wnt sensitive in all tissues, thus reducing tissue specificity.
  • the parameters describing the probabilistic relationships between the network elements have to be determined. Furthermore, in case of discrete states of the input measurements, thresholds have to be set that describe how to do the discretization.
  • Bayesian networks are trained using a representative set of training samples, of which preferably all states of all network nodes are known.
  • training samples preferably all states of all network nodes are known.
  • available training sets consist of a limited number of samples, typically from one type of cancer only.
  • the (unconditional) probability of being in state "absent” and “present” is given by the expected occurrence on a large set of samples.
  • the conditional probabilities are set as in Figure 1. If the TF element is "absent”, it is most likely that the target gene is "down”, hence a probability of 0.95 is chosen for this, and a probability of 0.05 for the target gene being "up". The latter (non-zero) probability is to account for the (rare) possibility that the target gene is regulated by other factors or accidentally observed “up” (e.g. because of measurement noise). If the TF element is "present”, then with a fair probability of 0.70 the target gene is "up”, and with a probability of 0.30 the target gene is "down”.
  • conditional probability tables describing the probabilities of a probeset to have a "low” or “high” intensity depending on the "down” or “up” state of the respective target gene is done by counting the number of "on” and “off samples with an intensity value of the probeset below and above the respective threshold. This is known in the literature as the frequentist approach. A dummy count is added to each group to prevent entries in the conditional probability tables with a value of zero, to prevent extreme behavior of the Bayesian network.
  • the Bayesian network After the Bayesian network has been trained, it can be applied on a test sample as follows, considering the Bayesian network of Figure 6, and assuming the microarray measurements relating to the probesets are available.
  • the first step is to discretize the input measurements, by comparing each probeset's intensity in the test sample to the respective threshold as described above. This comparison may be done in a hard way, setting each probeset to either "low” or “high” intensity (called 'hard evidence'), or it can be done in a soft way, assuming some uncertainty (noise) in the measurement, setting for each probeset a probability of being "low” or "high” (called 'soft evidence'). For instance, the soft evidence of a probeset with an intensity just below the threshold may be a probability of 0.8 of being "low” and a probability of 0.2 of being "high”, based on a suitable estimate of the noise and the difference to the threshold.
  • this hard or soft evidence is supplied to a suitable inference engine for Bayesian networks, for instance based on a junction tree algorithm (see (Neapolitan, 2004)).
  • a suitable inference engine for Bayesian networks, for instance based on a junction tree algorithm (see (Neapolitan, 2004)).
  • Such an engine can then infer the updated probability of the TF element being "absent” or “present”, given the provided evidence.
  • the inferred probability of the TF element being "present” is then interpreted as the estimated probability that the respective pathway is active.
  • the training of the Bayesian network models of the Wnt, ER, Hedgehog and AR pathways is done using public data available on the Gene Expression Omnibus (accessible at http://www.ncbi.nlm.nih.gov/geo/, cf. above).
  • the Wnt Bayesian network was exemplary trained using 32 normal colon samples considered to have an inactive Wnt pathway and 32 confirmed adenoma samples known to have an active Wnt pathway (GSE8671 data set).
  • the Bayesian network model of the ER pathway was exemplary trained using 4 estrogen-deprived MCF7 samples, known to have an inactive ER pathway, and 4 estrogen-stimulated MCF7 samples, regarded to have an active ER pathway, from the GSE8597 data set also accessible at the Gene expression Omnibus.
  • the Bayesian network model of the Hedgehog pathway was exemplary trained using 15 basal cell carcinoma samples confirmed to have an active Hedgehog pathway and 4 normal skin cells samples representing samples with an inactive Hedgehog pathway available in the GSE7553 data set.
  • the Bayesian network model of the AR pathway was exemplary trained using 3 samples with positive AR activity, LNCaP cell lines stimulated with
  • DHT Dihydrotestosterone
  • a potent AR pathway activator a potent AR pathway activator
  • 3 non-stimulated LNCaP cell lines representing the inactive AR pathway case.
  • the trained Bayesian network models of the Wnt and ER pathway were used to predict the pathway activities in similar samples (colon samples and MCF7 breast cancer cell line for the Wnt and ER Bayesian network, respectively) not used in the training procedure as described herein (no appropriate data set for the Hedgehog Bayesian network was found).
  • the predicted pathway activities of the vast majority of the samples should be in line with the clinically expected pathway activities for the model to be validated.
  • Figure 35 shows the predicted Wnt activities, depicted as the logit of P(Wnt on) on the vertical axis, for the samples, illustrated by the bars on the horizontal axis, of the colon samples grouped by classification, indicated by the bar's color, in the GSE20916 data set. All normal colon samples are rightfully predicted to have an inactive pathway
  • Bayesian network models to diagnose the activity of a cellular signaling pathway.
  • the Bayesian networks of the Wnt, ER, Hedgehog and AR pathway constructed using a node for the transcription factor presence, a layer of nodes representing the target genes' mRNA and a layer of nodes representing the probesets' intensities corresponding to the target genes (Table 1, Table 2, Table 3 and Table 4), analogous to Figure 6 described herein, and trained as described herein, were used to predict the pathways activity as "on", that is active, or "off, that is inactive, in various, previously not used for training, data sets to infer how well the inference component operates.
  • the predicted pathway activity scores are correlated with clinical knowledge. Result summaries for a selection of the test runs are shown in Figures 21 seq.
  • Figure 21 shows results for Wnt activity tests on colon samples data set
  • the Bayesian network model yielded high values of P(Wnt On) for the adenoma samples, and low values for normal samples, which corresponds with the
  • the Bayesian network model used in the experiments reported herein was trained using the colon samples data set GSE8671. However, the Wnt pathway is present (albeit possibly inactive) in other cell types. It was therefore considered possible that the Bayesian network might be applicable to infer abnormally high Wnt pathway activity correlative with other types of cancers. The rationale for this is that, although the Bayesian network model was trained using colon samples, it is based on first principles of the operation of the Wnt pathway present (albeit possibly inactive) in other cell types. Figures 22 - 24 show some results investigating such "cross tissue type" inferences.
  • Figure 22 shows results for tests using the Bayesian network model trained using colon samples being applied to infer Wnt pathway activity in medulloblastoma samples (data set GSE 10327).
  • the samples included in this data set have been further characterized in several subsets, one of them being samples with the Wnt pathway being active.
  • the Bayesian network of the Wnt pathway predicts the group of Wnt-active samples having an active Wnt pathway, whereas the other samples where predicted correctly to have an inactive Wnt pathway.
  • the test results using the Wnt Bayesian network model in a data set containing liver cancer samples (GSE9843) is shown in Figure 23.
  • the samples are grouped by the following a priori annotations assigned by the GSE9843 data set:
  • Proliferation suggests a state of rapid cellular multiplication. Such a state may be associated with abnormally high Wnt pathway activity, but may also be associated with numerous other possible causes of cell proliferation. Accordingly, about one in five of these samples having abnormally high Wnt pathway activity is not an unreasonable result.
  • the CTNNB1 gene encodes the beta-catenin protein, which is a regulatory protein of the Wnt pathway, and activating mutations in this gene cause abnormal Wnt activation.
  • a correlation between the "CTNNB1" group and high Wnt pathway activity is conform expectation.
  • Figure 24 depicts the test results of the Wnt Bayesian network model described herein for a set of breast cancer samples.
  • three groups of breast cancer cell lines are tested: one group for which the Wnt pathway is a priori known to be operating at an abnormally high level (Wnt on-group); one group for which the Wnt pathway is a priori known to not be operating at an abnormally high level (Wnt off-group); and another group for which the Wnt pathway activity is not a priori known (Unknown group); in addition there is also one sample that is suspected to have a low level of Wnt activation (Wnt suspicious), although there is a conflicting report in the literature that it may have an active Wnt pathway (but this is a minority report; more papers report an inactive Wnt pathway).
  • the correlation of the inferences provided by the Bayesian network with the a priori knowledge is strong for the Wnt on- and off-groups.
  • the rightmost sample of the graph shows an inference that corresponds to most reports in the literature stating that the Wnt pathway is off.
  • the Bayesian network infers low activity for the Wnt pathway except for one instance for which P(Wnt On)>0.5; literature shows that this cell line has a high expression of the co-receptor LRP6, which may explain that the Wnt pathway is on.
  • Figure 25 shows the results for the same data set of breast cancer cell lines but now tested for ER activity using the ER Bayesian network trained using MCF7 breast cancer cell lines as described herein.
  • the samples a priori known to have an active Wnt pathway were predicted to have an inactive ER pathway, which is not surprising since the Wnt pathway is already driving the rapid cell multiplication.
  • ER positive samples on the other hand are found amongst the Wnt off-samples and the unknown-samples. In view of Figure 24, this is not surprising.
  • the test results of the predictions of the ER Bayesian network trained on breast cancer cell lines for a set of cancer samples are shown in Figure 26.
  • the breast cancer samples were subdivided in the well-known classifications: Luminal A (LumA), Luminal B (LumB), Human Epidermal Growth Factor Receptor 2 positive (HER2), and basal breast cancer subtype.
  • Tissue samples in the luminal A and luminal B subtypes are known to express ER. It is also in these subtypes that most samples are predicted to have a high ER pathway activity.
  • the samples that are classified to be of the basal subtype are known to have no or low expression of ER, which nicely correlates with no active ER pathway predicted in the basal group samples.
  • the ER Bayesian network model constructed and trained as described herein is used to predict the ER pathway activity in a large panel of cell lines of various cancers, the results are shown in Figure 27. As expected only ER active predicted samples were found in the breast cancer cell lines. All other types of cancer cell lines were predicted to have an inactive ER pathway, which is as expected.
  • the Bayesian network model constructed and trained for the Hedgehog pathway as described herein is used to predict the activity of the Hedgehog pathway for cell lines of various cancer types in the GSE34211-data set.
  • the Hedgehog activity predictions are shown in Figure 28.
  • the highest fractions of positive predicted Hedgehog activity are found in the central nervous system (CNS), skin, endometrium, and uterus cancer types, which is in good agreement with literature knowledge regarding Hedgehog- dependent cell proliferation in these cell types.
  • Figure 29 shows the predicted Hedgehog activity of the medulloblastoma samples (GSE 10327) that was already analyzed using the Wnt Bayesian network model as described herein.
  • the medulloblastoma samples have been characterized in subclasses, with one of them having an active Hedgehog signaling pathway (identifier: SHH). All of the samples in the SHH subtype are predicted to have an active Hedgehog signaling.
  • the medulloblastoma samples in the Wnt-subtype were also predicted to have an active Hedgehog pathway. This is in agreement with clinical evidence showing that often both pathways are active in these tumors. Nevertheless, the Wnt Bayesian network was clearly able to correctly predict Wnt activity only in the Wnt-subtype.
  • the predicted Hedgehog activity in the GSE 12276 breast cancer samples, earlier used to predict the ER activity using the ER Bayesian network model, using the Hedgehog Bayesian network model is shown in Figure 30.
  • the Hedgehog pathway is predicted to be active in a fraction of the samples of every subtype. This seems odd, but matched to the ER pathway prediction shown in Figure 26 one can see that Hedgehog activity is only predicted in samples not having an active ER pathway. This is in good agreement with the hypothesis that uncontrolled cell proliferation in (breast) tissue can be driven by different signaling pathways.
  • the test results for various cancerous tissue samples and cells presented in Figures 21 - 30 strongly suggest that the Bayesian networks of the Wnt, ER and Hedgehog models trained on tissue/pathway specific samples are applicable to analysis of samples of other types of tissue. This can enable cellular signaling pathway analysis to be applied "cross tissue type".
  • the CDS system 10 (as described herein) is readily applied to assess pathway activity in a range of tissue types other than the tissue type of the samples used to train the Bayesian network model 40 (see e.g. Figure 20 which shows diagrammatically a clinical decision support (CDS) system configured to assess one or more cellular signaling pathways as disclosed herein (exemplary shown for Wnt pathway)).
  • CDS clinical decision support
  • inference components 40, 44, 46, 48 indicate the tissue under analysis exhibits abnormally high Wnt, ER or Hedgehog pathway activity, but no tissue specific drug is available, a general Wnt, ER or Hedgehog pathway suppression drug, or a malfunction specific drug, may be considered by the physician based on the recommendation 28 or the recommendation 26, respectively, as provided by the CDS system 10.
  • the Bayesian network models may optionally be updated or adapted to maximize its applicability to the specific tissue type under analysis (e.g., breast tissue or liver tissue).
  • tissue type under analysis e.g., breast tissue or liver tissue.
  • Such updating or adaptation could, for example, entail adjusting the conditional probabilities based on clinical studies of the tissue type under analysis or enrich the evidence curated target gene list, described herein, with tissue specific target genes of the pathway(s) under investigation.
  • nodes might be added or removed to better tune the Bayesian network model to the tissue under analysis.
  • different Bayesian network models may be trained ab initio using different training sets for the different tissue types.
  • the results of Figures 21 - 30 illustrate the ability of the process described herein to develop and train Bayesian network models using evidence curated target gene lists of pathways other than Wnt, ER and Hedgehog to predict and diagnose the pathway's activity.
  • the test results of the AR Bayesian network model constructed and trained as described herein was exemplary used to predict the AR activity in LNCaP prostate cancer cell lines treated with different treatment regimes (GSE7708) (see Figure 38).
  • GSE7708 treatment regimes
  • LNCaP cells not stimulated with DHT results in a predicted inactive AR pathway
  • LNCaP cells treated with Bicalutamide, an anti-androgen drug to have an inhibited AR pathway.
  • the trained Bayesian network of the AR pathway as described herein was also used to predict the probability the AR pathway is active in prostate cancer samples from the GSE17951 data set (results are shown in Figure 39). The majority of the prostate biopsies and tumors were not unexpectedly predicted to have a higher probability of AR activity compared to the controls samples.
  • the AR Bayesian network model was also applied to a cross-tissue test, viz. the breast cancer samples included in the GSE 12276 data set. Results for this test are shown in Figure 40. A small fraction of the samples, found in every subgroup, are predicted to have an active pathway, whereas the vast majority of the samples had an inactive AR pathway. Remarkably the highest percentage of samples with an active AR pathway are found in the HER2-subgroup, which is not unexpected as it is known from literature that there is crosstalk between the HER2 and AR pathway and the AR pathway can also be induced by HER2-signaling.
  • Example 7 Prognosis based on pathway activity
  • breast cancer patients from the GSE 12276 data set, that were identified having an active ER pathway but not having an active Wnt or Hedgehog pathway using the Bayesian network models described herein had a better prognosis than patients identified having either an active Hedgehog or Wnt pathway or both, as illustrated by the Kaplan-Meier plot in Figure 37.
  • Example 8 Therapy planning, prediction of drug efficacy, prediction of adverse effects and monitoring of drug efficacy
  • the following exemplary illustrates how to use the probabilistic models, in particular Bayesian network models, for therapy planning, prediction of drug efficacy, monitoring of drug efficacy and related activities.
  • the Bayesian network model of the ER pathway constructed using a node for the transcription factor presence, a layer of nodes representing the target genes' mRNA levels (Table 2) and a layer of nodes representing the probesets' intensities corresponding to the target genes (Table 2), analogous to Figure 6 described herein, and trained as described herein, were used to predict the pathway activity.
  • the pathway activity is subsequently demonstrated to be correlated with drug efficacy or monitoring drug efficacy. Result summaries are shown in Figures 31 and 32.
  • Tamoxifen is a drug currently used for the treatment of ER+ (estrogen receptor positive) breast cancer. It acts as a partial antagonist of the estrogen receptor inhibiting the uncontrolled cell proliferation which is thought to be induced by ER signaling.
  • ER+ estrogen receptor positive
  • Tamoxifen resistance Unfortunately, not every breast cancer responds to treatment with Tamoxifen, despite the demonstration of the presence of ER protein in cancer cells by routine histopathology analysis of cancer tissue slides. Many studies have been conducted to investigate this so-called Tamoxifen resistance.
  • the publicly available GSE21618 data set is the result of one of such study and contains microarray data of Tamoxifen resistant and wildtype MCF7 cell lines under different treatment regimes.
  • the ER Bayesian network model constructed and trained as described herein is used to analyze the Tamoxifen resistant and MCF7 cell lines under different treatment regimes, the results are depicted in Figure 31.
  • the control Tamoxifen resistant cell line is predicted to have an inactive ER pathway for every time point after Tamoxifen addition (1, 2, 3, 6, 12, 24, and 48 h). It is not surprising that treatment of the Tamoxifen resistant cell line, that is insensitive to Tamoxifen treatment, with Tamoxifen, indicated by TamR.Tam, is ineffective, which is also illustrated by the predicted inactivity of the ER pathway for this group over the same time points. According to analysis of the Tamoxifen resistant cell line (TamR.Ctrl) the driving force of the uncontrolled cell proliferation is not due to active ER signaling; therefore treating it with an ER antagonist will not inhibit cell proliferation. This illustrates that treatment with Tamoxifen is not recommended in case of a negative predicted ER pathway activity.
  • the wild type MCF7 cell line known to be Tamoxifen sensitive, treated with 17beta-estradiol (wtl .E2) slowly reacts to the hormone treatment which is visible in the increasing ER positive activity predictions.
  • Treating such a cell line with aromatase inhibitors that are known to inhibit estrogen production will inhibit the ER pathway which is illustrated by the decreasing ER pathway prediction in time.
  • Supporting this are the ER pathway predictions made based on the microarray data from MCF7 samples treated with estrogen for increasing time in the GSE11324 data set, results shown in Figure 32.
  • the above mentioned illustrates the ability of the probabilistic models, in particular the Bayesian network models, to be used for therapy planning, drug efficacy prediction, and monitoring of drug efficacy. However it is to be understood, the same methodology would also apply to predict and monitor adverse effects.
  • a pathway model can be used in drug development to assess the effectiveness of various putative compounds. For instance, when screening many compounds for a possible effect on a certain pathway in a cancer cell line, the respective pathway model can be used to determine whether the activity of the pathway goes up or down after application of the compound or not. Often, this check is done using only one or a few of putative markers of the pathway's activity, which increases the chance of ineffective monitoring of the treatment effect. Furthermore, in follow-up studies on animal or patient subjects, the pathway models can be used similarly to assess the effectiveness of candidate drugs, and to determine an optimal dose to maximally impact pathway activity.
  • RNA/DNA sequences of the disclosed target genes can then be used to determine which primers and probes to select on such a platform.
  • Validation of such a dedicated assay can be done by using the microarray- based Bayesian networks as a reference model, and verifying whether the developed assay gives similar results on a set of validation samples. Next to a dedicated assay, this can also be done to build and calibrate similar Bayesian network models using niRNA-sequencing data as input measurements.
  • Example 11 Pathway research and cancer pathophysiology research
  • Bayesian network models can be employed in (clinical) pathway research, that is research interested to find out which pathways are involved in certain diseases, which can be followed up for more detailed research, e.g. to link mutations in signaling proteins to changes in pathway activation (measured with the model). This is relevant to investigate the initiation, growth and evolution and metastasis of specific cancers (the pathophysiology).
  • Bayesian network models of the Wnt, ER, Hedgehog and AR pathway constructed using a node for the transcription factor presence, a layer of nodes representing the target genes' mRNA levels (Table 1, Table 2, Table 3 and Table 4) and a layer of nodes representing the probesets' intensities corresponding to the target genes (Table 1, Table 2, Table 3 and Table 4), analogous to Figure 6 described herein, and trained as described herein, were used to predict the pathway activity of a data set consisting of breast cancer samples (GSE12276).
  • the researcher is interested in looking into the cellular signaling pathway or pathways and the specific deregulation(s) that drive(s) the uncontrolled cell proliferation.
  • the researcher can analyze the microarray data using the above mentioned probabilistic models, in particular the Bayesian network models, to find which pathways are presumably the cause of uncontrolled cell proliferation. Shown in Figure 33 and Figure 34 one can see an illustration of such an analysis for the case of Wnt, ER and Hedgehog activity (basal and luminal A samples of the GSE 12276 data set). Subsequently, the researcher can search in more detail to find the exact cause of pathway deregulation.
  • the basal samples are known to have triple negative receptor status (ER, PR and HER2), therefore it is not surprising to see that all samples are predicted to have an inactive ER pathway.
  • some of the samples are predicted to have either the Wnt or Hedgehog or both active as shown in Figure 34. These predicted pathway activities persuade the researcher to investigate these samples in more detail for e.g. known mutations or other known deregulations in the Wnt and/or Hedgehog pathways.
  • FIG. 33 Another example is given in Figure 33, where the Wnt, ER and Hedgehog activities in the luminal A samples of the GSE 12276 data set are illustrated.
  • Luminal A samples are known to express ER, however this does not necessarily mean the cancerous properties are due to active ER signaling. From the predicted pathway activities one can infer that less than half of the samples have an active ER signaling. However, some of the samples that do not have an active ER signaling are found to have an active Wnt and/or Hedgehog pathway. This might give rise for the researcher to investigate these samples in closer details for defects in the Wnt and/or Hedgehog signaling pathway, respectively.
  • the illustrations described herein indicate the ability of trained Bayesian network models (as described above) to support the process of finding the cause of uncontrolled cell proliferation in a more directed method.
  • the predicted pathway activities can pinpoint the possible pathways for the cell proliferation, which can be followed up for more detailed research, e.g. to link mutations in signaling proteins or other known deregulations to changes in activation (as measured with the model).
  • the process to develop and train a Bayesian network of cellular signaling pathways can be used to construct a Bayesian network model for other pathways that could also be employed in connection with the present invention.
  • Example 12 Enrollment of subject in a clinical trial based on predicted activity
  • a candidate drug is developed to, for instance, block the activity of a certain pathway that drives tumor growth, and this drug is going into clinical trial, then a proper selection of the subjects to enroll in such a trial is essential to prove potential effectiveness of the drug.
  • patients that do not have the respective pathway activated in their tumors should be excluded from the trial, as it is obvious that the drug cannot be effective if the pathway is not activated in the first place.
  • a pathway model that can predict pathway activity can be used as a selection tool, to only select those patients that are predicted to have the respective pathway activated.
  • Example 13 Selection of subsequent test(s) to be performed
  • PLA proximity ligation assay
  • the pathway predicted to be deregulated is analyzed in more detail with respect to the signaling cascade. For instance, one may analyze key proteins in this pathway to determine whether there are mutations in the DNA regions encoding for their respective genes, or one may test for the abundance of these proteins to see whether they are higher or lower than normal. Such tests may indicate what the root cause is behind the deregulation of the pathway, and give insights on which available drugs could be used to reduce activity of the pathway. These tests are selected to confirm the activity of the pathway as identified using the Bayesian model. However selection of companion diagnostic tests is also possible. After identification of the pathway using the model, for targeted therapy choice only those companion diagnostics tests need to be performed (the selection), which are applicable to the identified pathway.
  • a clinical decision support (CDS) system 10 is implemented as a suitably configured computer 12.
  • the computer 12 may be configured to operate as the CDS system 10 by executing suitable software, firmware, or other instructions stored on a non-transitory storage medium (not shown) such as a hard drive or other magnetic storage medium, an optical disk or other optical storage medium, a random access memory (RAM), read-only memory (ROM), flash memory, or other electronic storage medium, a network server, or so forth.
  • a non-transitory storage medium such as a hard drive or other magnetic storage medium, an optical disk or other optical storage medium, a random access memory (RAM), read-only memory (ROM), flash memory, or other electronic storage medium, a network server, or so forth.
  • the illustrative CDS system 10 is embodied by the illustrative computer 12, more generally the CDS system may be embodied by a digital processing device or an apparatus comprising a digital processor configured to perform clinical decision support methods as set forth herein.
  • the digital processing device may be a handheld device (e.g., a personal data assistant or smartphone running a CDS application), a notebook computer, a desktop computer, a tablet computer or device, a remote network server, or so forth.
  • the computer 12 or other digital processing device typically includes or is operatively connected with a display device 14 via which information including clinical decision support recommendations are displayed to medical personnel.
  • the computer 12 or other digital processing device typically also includes or is operatively connected with one or more user input devices, such as an illustrative keyboard 16, or a mouse, trackball, trackpad, touch-sensitive screen (possibly integrated with the display device 14), or other pointer-based user input device, via which medical personnel can input information such as operational commands for controlling the CDS system 10, data for use by the CDS system 10, or so forth.
  • user input devices such as an illustrative keyboard 16, or a mouse, trackball, trackpad, touch-sensitive screen (possibly integrated with the display device 14), or other pointer-based user input device, via which medical personnel can input information such as operational commands for controlling the CDS system 10, data for use by the CDS system 10, or so forth.
  • the CDS system 10 receives as input information pertaining to a medical subject (e.g., a hospital patient, or an outpatient being treated by an oncologist, physician, or other medical personnel, or a person undergoing cancer screening or some other medical diagnosis who is known or suspected to have a certain type of cancer such as colon cancer, breast cancer, or liver cancer, or so forth).
  • the CDS system 10 applies various data analysis algorithms to this input information in order to generate clinical decision support recommendations that are presented to medical personnel via the display device 14 (or via a voice synthesizer or other device providing human-perceptible output). In some embodiments, these algorithms may include applying a clinical guideline to the patient.
  • a clinical guideline is a stored set of standard or “canonical” treatment recommendations, typically constructed based on recommendations of a panel of medical experts and optionally formatted in the form of a clinical "flowchart" to facilitate navigating through the clinical guideline.
  • the data processing algorithms of the CDS 10 may additionally or alternatively include various diagnostic or clinical test algorithms that are performed on input information to extract clinical decision recommendations, such as machine learning methods disclosed herein.
  • the CDS data analysis algorithms include one or more diagnostic or clinical test algorithms that are performed on input genomic and/or proteomic information acquired by one or more medical laboratories 18. These laboratories may be variously located "on-site", that is, at the hospital or other location where the medical subject is undergoing medical examination and/or treatment, or "off-site", e.g.
  • tissue of the medical subject that has been extracted from the medical subject (e.g., a sample obtained from a breast lesion, or from a colon of a medical subject known or suspected of having colon cancer, or from a liver of a medical subject known or suspected of having liver cancer, or so forth, via a biopsy procedure or other sample extraction procedure).
  • tissue of which a sample is extracted may also be metastatic tissue, e.g. (suspected) malignant tissue originating from the colon, breast, liver, or other organ that has spread outside of the colon, breast, liver, or other organ.
  • the tissue sample may be circulating tumor cells, that is, tumor cells that have entered the bloodstream and may be extracted as the extracted tissue sample using suitable isolation techniques.
  • the extracted sample is processed by the laboratory to generate genomic or proteomic information.
  • the extracted sample may be processed using a microarray (also variously referred to in the art as a gene chip, DNA chip, biochip, or so forth) or by quantitative polymerase chain reaction (qPCR) processing to measure probative genomic or proteomic information such as expression levels of genes of interest, for example in the form of a level of messenger ribonucleic acid (mR A) that is transcribed from the gene, or a level of a protein that is translated from the mRNA transcribed from the gene.
  • mR A messenger ribonucleic acid
  • the extracted sample may be processed by a gene sequencing laboratory to generate sequences for deoxyribonucleic acid (DNA), or to generate an RNA sequence, copy number variation, or so forth.
  • RNA sequence may be processed by a gene sequencing laboratory to generate sequences for deoxyribonucleic acid (DNA), or to generate an RNA sequence, copy number variation, or so forth.
  • Other contemplated measurement approaches include immunohistochemistry (IHC), cytology, fluorescence in situ hybridization (FISH), proximity ligation assay or so forth, performed on a pathology slide.
  • IHC immunohistochemistry
  • FISH fluorescence in situ hybridization
  • proximity ligation assay or so forth
  • spectrometry includes methylation information. Various combinations of such genomic and/or proteomic measurements may also be performed.
  • the medical laboratories 18 perform a number of standardized data acquisitions on the extracted sample of the tissue of the medical subject, so as to generate a large quantity of genomic and/or proteomic data.
  • the standardized data acquisition techniques may generate an (optionally aligned) DNA sequence for one or more chromosomes or chromosome portions, or for the entire genome of the tissue.
  • Applying a standard microarray can generate thousands or tens of thousands of data items such as expression levels for a large number of genes, various methylation data, and so forth.
  • This plethora of genomic and/or proteomic data, or selected portions thereof, are input to the CDS system 10 to be processed so as to develop clinically useful information for formulating clinical decision support recommendations.
  • the disclosed CDS systems and related methods relate to processing of genomic and/or proteomic data to assess activity of various cellular signaling pathways.
  • the disclosed CDS systems may optionally further include diverse additional capabilities, such as generating clinical decision support recommendations in accordance with stored clinical guidelines based on various patient data such as vital sign monitoring data, patient history data, patient demographic data (e.g., gender, age, or so forth), patient medical imaging data, or so forth.
  • the capabilities of the CDS system 10 may be limited to only performing genomic and/or proteomic data analyses to assess cellular signaling pathways as disclosed herein.
  • the CDS system 10 infers activity of a cellular signaling pathway in the tissue of the medical subject based at least on, but not restricted to, expression levels of target genes of the cellular signaling pathway measured in the extracted sample, and determines whether the cellular signaling pathway is operating abnormally in the tissue of the medical subject based on this inferred activity.
  • Examples disclosed herein relate to the Wnt, ER, AR and Hedgehog pathways as illustrative cellular signaling pathways. These pathways are of interest in various areas of oncology because loss of regulation of the pathways can be a cause of proliferation of a cancer. There are about 10-15 relevant signaling pathways, and each cancer is driven by in principle one dominant pathway being deregulated.
  • these pathways regulate cell proliferation, and consequentially a loss of regulation of these pathways in cancer cells can lead to the pathway being "always on” thus accelerating the proliferation of cancer cells, which in turn manifests as a growth, invasion or metastasis (spread) of the cancer.
  • Measurement of mRNA expression levels of genes that encode for regulatory proteins of the cellular signaling pathway is an indirect measure of the regulatory protein expression level and may or may not correlate strongly with the actual regulatory protein expression level (much less with the overall activity of the cellular signaling pathway).
  • the cellular signaling pathway directly regulates the transcription of the target genes - hence, the expression levels of mR A transcribed from the target genes is a direct result of this regulatory activity.
  • the CDS system 10 infers activity of the cellular signaling pathway (e.g., the Wnt, ER, AR and Hedgehog pathways) based at least on expression levels of target genes (mRNA or protein level as a surrogate measurement) of the cellular signaling pathway. This ensures that the CDS system 10 infers the activity of the pathway based on direct information provided by the measured expression levels of the target genes.
  • the cellular signaling pathway e.g., the Wnt, ER, AR and Hedgehog pathways
  • target genes mRNA or protein level as a surrogate measurement
  • the measured expression levels 20 of target genes of the pathways are not especially informative as to why the pathways are operating abnormally (if indeed that is the case). Said another way, the measured expression levels 20 of target genes of a pathway can indicate that the pathway is operating abnormally, but do not indicate what portion of the pathway is malfunctioning (e.g., lacks sufficient regulation) in order to cause the overall pathway to operate abnormally.
  • the CDS system 10 detects abnormal activity of a particular pathway, the CDS system 10 then optionally makes use of other information provided by the medical laboratories 18 for the extracted sample, such as aligned genetic sequences 22 and/or measured expression level(s) for one or more regulatory genes of the pathway 24, or select the diagnostic test to be performed next in order to assess what portion of the pathway is malfunctioning.
  • this optional assessment of why the pathway is malfunctioning is performed only if the analysis of the measured expression levels 20 of target genes of the pathway indicates that the pathway is operating abnormally. In other embodiments, this assessment is integrated into the probabilistic analysis of the cellular signaling pathway described herein.
  • the additional information enables the CDS system 10 to recommend prescribing a drug targeting for the specific malfunction (recommendation 26 shown in Figure 20). If no specific pathway malfunction is identified (either because the optional additional assessment is not performed or because that assessment fails to identify any particular portion of the pathway that is malfunctioning), then the CDS system 10 can provide a default recommendation 28 recommending the prescription of a general suppression drug for this particular pathway (assuming that the abnormal pathway activity is overly high activity).
  • Example 16 A kit and analysis tools to measure pathway activity
  • the set of target genes which are found to best indicate specific pathway activity can be translated into a multiplex quantitative PCR assay to be performed on a tissue or cell sample.
  • development of a standardized test kit is required, which needs to be clinically validated in clinical trials to obtain regulatory approval.

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Chemical & Material Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Biotechnology (AREA)
  • Biophysics (AREA)
  • Medical Informatics (AREA)
  • Theoretical Computer Science (AREA)
  • Molecular Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Analytical Chemistry (AREA)
  • Genetics & Genomics (AREA)
  • Organic Chemistry (AREA)
  • Zoology (AREA)
  • Wood Science & Technology (AREA)
  • Data Mining & Analysis (AREA)
  • Immunology (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Physiology (AREA)
  • Evolutionary Computation (AREA)
  • Biochemistry (AREA)
  • Microbiology (AREA)
  • Bioethics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Databases & Information Systems (AREA)
  • Epidemiology (AREA)
  • Public Health (AREA)
  • Pathology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Hospice & Palliative Care (AREA)
  • Oncology (AREA)
  • Mathematical Physics (AREA)

Abstract

The present application mainly relates to specific methods for inferring activity of one or more cellular signaling pathway(s) in tissue of a medical subject based at least on the expression level(s) of one or more target gene(s) of the cellular signaling pathway(s) measured in an extracted sample of the tissue of the medical subject, an apparatus comprising a digital compressor configured to perform such methods and a non- transitory storage medium storing instructions that are executable by a digital processing device to perform such methods.

Description

ASSESSMENT OF CELLULAR SIGNALING PATHWAY ACTIVITY USING
PROBABILISTIC MODELING OF TARGET GENE EXPRESSION
The subject matter described herein mainly relates to bio informatics, genomic processing arts, proteomic processing arts, and related arts.
Genomic and proteomic analyses have substantial realized and potential promise for clinical application in medical fields such as oncology, where various cancers are known to be associated with specific combinations of genomic mutations/ variations and/or high or low expression levels for specific genes, which play a role in growth and evolution of cancer, e.g. cell proliferation and metastasis. For example, the Wnt signaling pathway affects regulation of cell proliferation, and is highly regulated. High Wnt pathway activity due to loss of regulation has been correlated to cancer, among which with malignant colon tumors. While not being limited to any particular theory of operation, it is believed that deregulation of the Wnt pathway in malignant colon cells leads to high Wnt pathway activity that in turn causes cell proliferation of the malignant colon cells, i.e. spread of colon cancer. On the other hand, abnormally low pathway activity might also be of interest, for example in the case of osteoporosis.
Technologies for acquiring genomic and proteomic data have become readily available in clinical settings. For example, measurements by microarrays are routinely employed to assess gene expression levels, protein levels, methylation, and so forth. Automated gene sequencing enables cost-effective identification of genetic variations in DNA and mRNA. Quantitative assessment of mRNA levels during gene sequencing holds promise as yet another clinical tool for assessing gene expression levels.
In spite of (or, perhaps, because of) these advances, clinical application of genomic and proteomic analyses faces a substantial hurdle - data overload. For example, the number of identifiable mutations in a single clinical sample can number in the hundreds of thousands or more. Most of these mutations are so called bystander mutations without specific contribution to cancer growth, and only a few do contribute to cancer growth and functional evolution, and these present the targets for effective treatment. A single microarray can generate gene expression levels for tens of thousands of genes. Processing these large quantities of data to identify clinically useful information, like for example in the application of choosing the right therapy, is difficult.
One approach is to limit the analysis to a few canonical or standardized tests, such as tests approved by the U.S. Food and Drug Administration (FDA). In such an approach, a specific indicator or combination of indicators (e.g., mutations and/or specified high or low gene expression levels) is detected in order to test "positive" for the indicated disease condition (e.g., a particular type of cancer). The canonical test is supported by clinical studies that have shown strong correlation with the disease condition or with treatment efficacy. This approach is useful only for those clinical conditions for which a canonical test has been developed, e.g. specific diagnosis of a disease, or predicting response to a drug in a specific cancer type at a specific stage, and is also rigid as it is only applicable for the canonical conditions.
Another approach is based on identification of functionally related groups of genomic or proteomic indicators. For example, the Wnt pathway comprises a cascade of proteomic reactions. Major components of this chain include (but are not limited to) binding of the Wnt signaling protein to a frizzled surface receptor of the cell which causes activation of proteins of the disheveled family of proteins which in turn impact the level of transcription agents such as -catenin/TCF4 based protein complexes in the cell nucleus.
These transcription agents, in turn, control transcription of target mR A molecules that in turn are translated into target proteins of the Wnt pathway. Clinical studies have shown some correlations between regulatory proteins of the Wnt pathway and the activity of the
Wnt pathway.
However, applying such clinical study results to the diagnosis and clinical evaluation of a specific patient is difficult due to the complexity of signaling pathways, e.g. the Wnt pathway. As a simple example, measurement of the expression level of a protein that is "upstream" in the Wnt pathway may fail to detect abnormal behavior of a protein that is "downstream" in the Wnt pathway. It is believed that the Wnt pathway includes numerous feedback mechanisms and the simplified concept of "upstream" and
"downstream" may be inapplicable for a substantial portion of the Wnt pathway; more generally, abnormal behavior in one portion of the protein cascade comprising the Wnt pathway may have more or less effect on other portions of the protein cascade, and on the activity of the Wnt pathway as a whole. Still further, in some clinical studies protein expression levels for regulatory proteins of the signaling cascade are assessed by measuring mR A expression levels of the genes that encode for the regulatory proteins. This is an indirect measurement that may not accurately assess the regulatory protein expression level, and hardly ever reflects the amount of active proteins (after a specific post-translational modification like phosphorylation).
The main problem underlying the present invention was thus to provide suitable methods and means for performing genomic and, respectively, proteomic analyses Specific aspects of the underlying problem as well as further objections in connection with the present invention become apparent when studying the description, the examples provided herein and, in particular, when studying the attached claims.
The present invention provides new and improved methods and apparatuses as disclosed herein.
In accordance with a main aspect of the present invention, the above problem is solved by a specific method for assessing cellular signaling pathway activity using probabilistic modeling of target gene expression, namely a method comprising:
inferring activity of one or more cellular signaling pathway(s) in tissue of a medical subject based at least on the expression level(s) (in particular on mRNA and/or protein level) of one or more target gene(s) of the cellular signaling pathway(s) measured in an extracted sample of the tissue of the medical subject, wherein the inferring comprises inferring activity of the cellular signaling pathway(s) in the tissue of the medical subject by evaluating at least a portion of a probabilistic model, preferably a Bayesian network, representing the cellular signaling pathway(s) for a set of inputs including at least the expression level(s) of the one or more target genes of the cellular signaling pathway(s) measured in the extracted sample of the tissue of the medical subject;
estimating a level in the tissue of the medical subject of at least one transcription factor (TF) element, the at least one TF element controlling transcription of the one or more target gene(s) of the cellular signaling pathway(s), the estimating being based at least in part on conditional probabilities relating the at least one TF element and the expression level(s) of the one or more target gene(s) of the cellular signaling pathway(s) measured in the extracted sample of the tissue of the medical subject; and
inferring activity of the cellular signaling pathway(s) based on the estimated level in the tissue sample of the transcription factor; and determining whether the cellular signaling pathway(s) is/are operating abnormally in the tissue of the medical subject based on the inferred activity of the cellular signaling pathway(s) in the tissue of the medical subject;
wherein the inferring is performed by a digital processing device using the probabilistic model of the cellular signaling pathway(s).
The "target gene(s)" may be "direct target genes" and/or "indirect target genes" (as described herein).
Preferably the inferring comprises estimating a level in the tissue of the medical subject of at least one transcription factor (TF) element represented by a TF node of the probabilistic model, the TF element controlling transcription of the one or more target gene(s) of the cellular signaling pathway(s), the estimating being based at least in part on conditional probabilities of the probabilistic model relating the TF node and nodes in the probabilistic model representing the one or more target gene(s) of the cellular signaling pathway(s) measured in the extracted sample of the tissue of the medical subject.
The probabilistic model may be a Bayesian network model. Thus, according to a preferred embodiment the inferring is performed by using a Bayesian network comprising nodes representing information about the signaling pathway(s) and conditional probability relationships between connected nodes of the Bayesian network.
The cellular signaling pathway(s) may be a Wnt pathway, an ER (Estrogen Receptor) pathway, an AR (Androgen Receptor) pathway and/or a Hedgehog pathway.
Thus, according to a preferred embodiment the cellular signaling pathway(s) comprise(s) a Wnt pathway, an ER pathway, an AR pathway and/or a Hedgehog pathway.
Particularly suitable target genes are described in the following text passages as well as the examples below (see e.g. Tables 1 - 9).
Thus, according to a preferred embodiment the target gene(s) is/are selected from the group comprising or consisting of target genes listed in Table 1 or Table 6 (for Wnt pathway), target genes listed in Table 2, Table 5 or Table 7 (for ER pathway), target genes listed in Table 3 or Table 8 (for Hedgehog pathway) and target genes listed in Table 4 or Table 9 (for AR pathway).
Particularly preferred is a method wherein the inferring comprises:
inferring activity of a Wnt pathway in the tissue of the medical subject based at least on expression levels of one or more, preferably at least three, target gene(s) of the Wnt pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: KIAA1199, AXIN2, RNF43, TBX3, TDGF1, SOX9, ASCL2, IL8, SP5, ZNRF3, KLF6, CCND1, DEFA6 and FZD7.
Further preferred is a method, wherein the inferring is further based on expression levels of at least one target gene of the Wnt pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: NKD1, OAT, FAT1, LEF1, GLUL, REG1B, TCF7L2, COL18A1, BMP7, SLC1A2, ADRA2C, PPARG, DK 1, HNF1A and LECT2.
Particularly preferred is a method wherein the inferring (also) comprises: inferring activity of an ER pathway in the tissue of the medical subject based at least on expression levels of one or more, preferably at least three, target gene(s) of the ER pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: CDH26, SGK3, PGR, GREBl, CA12, XBP1, CELSR2, WISP2, DSCAM, ERBB2, CTSD, TFF1 and NRIPl .
Further preferred is a method, wherein the inferring is further based on expression levels of at least one target gene of the ER pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: AP1B1, ATP5J, COL18A1, COX7A2L, EBAG9, ESR1, HSPB1, IGFBP4, KRT19, MYC, NDUFV3, PISD, PREDM15, PTMA, RARA, SOD1 and TRIM25.
A method wherein the inferring (also) comprises
inferring activity of a Hedgehog pathway in the tissue of the medical subject based at least on expression levels of one or more, preferably at least three, target gene(s) of the Hedgehog pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: GLI1, PTCH1, PTCH2, IGFBP6, SPP1, CCND2, FST, FOXL1, CFLAR, TSC22D1, RAB34, S100A9, S100A7, MYCN, FOXM1, GLI3, TCEA2, FYN and CTSL1,
is also preferred.
Further preferred is a method, wherein the inferring is further based on expression levels of at least one target gene of the Hedgehog pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: BCL2, FOXA2, FOXF1, G19, HHIP, IL1R2, JAG2, JUP, MIF, MYLK, NKX2.2, NKX2.8, PITRM1 and TOM1. A method wherein the inferring (also) comprises
inferring activity of an AR pathway in the tissue of the medical subject based at least on expression levels of one or more, preferably at least three, target gene(s) of the AR pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: KLK2, PMEPA1, TMPRSS2, NKX3 1, ABCC4, KLK3, FKBP5, ELL2, UGT2B15, DHCR24, PPAP2A, NDRG1, LRIG1, CREB3L4, LCP1, GUCY1A3, AR and EAF2,
is also preferred.
Further preferred is a method, wherein the inferring is further based on expression levels of at least one target gene of the AR pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising or consisting of: APP, NTS, PLAU, CDKN1A, DRG1, FGF8, IGF1, PRKACB, PTPN1, SGK1 and TACC2.
Another aspect of the present invention relates to a method (as described herein), further comprising:
recommending prescribing a drug for the medical subject that corrects for abnormal operation of the cellular signaling pathway(s);
wherein the recommending is performed only if the cellular signaling pathway(s) is/are determined to be operating abnormally in the tissue of the medical subject based on the inferred activity of the cellular signaling pathway(s).
The present invention also relates to a method (as described herein) comprising:
inferring activity of a Wnt pathway in tissue of a medical subject based at least on expression levels of two, three or more target genes of a set of target genes of the Wnt pathway measured in an extracted sample of the tissue of the medical subject
and/or
inferring activity of an ER pathway in tissue of a medical subject based at least on expression levels of two, three or more target genes of a set of target genes of the ER pathway measured in an extracted sample of the tissue of the medical subject
and/or
inferring activity of a Hedgehog pathway in tissue of a medical subject based at least on expression levels of two, three or more target genes of a set of target genes of the Hedgehog pathway measured in an extracted sample of the tissue of the medical subject,
and/or
inferring activity of an AR pathway in tissue of a medical subject based at least on expression levels of two, three or more target genes of a set of target genes of the AR pathway measured in an extracted sample of the tissue of the medical subject.
Preferably,
the set of target genes of the Wnt pathway includes at least nine, preferably all target genes selected from the group comprising or consisting of: KIAAl 199, AXIN2, RNF43, TBX3, TDGF1, SOX9, ASCL2, IL8, SP5, ZNRF3, KLF6, CCND1, DEFA6 and FZD7,
and/or
the set of target genes of the ER pathway includes at least nine, preferably all target genes selected from the group comprising or consisting of: CDH26, SGK3, PGR, GREB1, CA12, XBP1, CELSR2, WISP2, DSCAM, ERBB2, CTSD, TFF1 and NRIPl, and/or
the set of target genes of the Hedgehog pathway includes at least nine, preferably all target genes selected from the group comprising or consisting of: GLI1, PTCH1, PTCH2, IGFBP6, SPP1, CCND2, FST, FOXL1, CFLAR, TSC22D1, RAB34, S100A9, S100A7, MYCN, FOXM1, GLI3, TCEA2, FYN and CTSL1,
and/or
the set of target genes of the AR pathway includes at least nine, preferably all target genes selected from the group comprising or consisting of: KLK2, PMEPA1, TMPRSS2, NKX3 1, ABCC4, KLK3, FKBP5, ELL2, UGT2B15, DHCR24, PPAP2A, NDRG1, LRIG1, CREB3L4, LCP1, GUCY1A3, AR and EAF2.
A method, wherein
the set of target genes of the Wnt pathway further includes at least one target gene selected from the group comprising or consisting of: NKD1, OAT, FAT1, LEF1, GLUL, REGIB, TCF7L2, COL18A1, BMP7, SLC1A2, ADRA2C, PPARG, DKK1, HNF1A and LECT2,
and/or the set of target genes of the ER pathway further includes at least one target gene selected from the group comprising or consisting of: AP1B1, ATP5J, COL18A1,
COX7A2L, EBAG9, ESR1, HSPB1, IGFBP4, KRT19, MYC, NDUFV3, PISD,
PREDM15, PTMA, RARA, SOD1 and TRIM25,
and/or
the set of target genes of the Hedgehog pathway further includes at least one target gene selected from the group comprising or consisting of: BCL2, FOXA2, FOXFl, G19, HHIP, IL1R2, JAG2, JUP, MIF, MYLK, NKX2.2, NKX2.8, PITRM1 and TOM1, and/or
the set of target genes of the AR pathway further includes at least one target gene selected from the group comprising or consisting of: APP, NTS, PLAU, CDK 1 A, DRG1, FGF8, IGF1, PRKACB, PTPN1, SGK1 and TACC2,
is particularly preferred.
The sample(s) to be used in accordance with the present invention can be, e.g., a sample obtained from a breast lesion, or from a colon of a medical subject known or suspected of having colon cancer, or from a liver of a medical subject known or suspected of having liver cancer, or so forth, preferably via a biopsy procedure or other sample extraction procedure. The tissue of which a sample is extracted may also be metastatic tissue, e.g. (suspected) malignant tissue originating from the colon, breast, liver, or other organ that has spread outside of the colon, breast, liver, or other organ. In some cases, the tissue sample may be circulating tumor cells, that is, tumor cells that have entered the bloodstream and may be extracted as the extracted tissue sample using suitable isolation techniques.
Another disclosed aspect of the present invention pertain the use of a non- transitory storage medium as described herein or a computer program as described herein for specific diagnosis of a disease or predicting response to a drug in a specific cancer type at a specific stage.
In accordance with another disclosed aspect, an apparatus comprising a digital processor configured to perform a method according to the invention as described herein. In accordance with another disclosed aspect, a non-transitory storage medium stores instructions that are executable by a digital processing device to perform a method according to the invention as described herein. The non-transitory storage medium may be a computer-readable storage medium, such as a hard drive or other magnetic storage medium, an optical disk or other optical storage medium, a random access memory (RAM), read-only memory (ROM), flash memory, or other electronic storage medium, a network server, or so forth. The digital processing device may be a handheld device (e.g., a personal data assistant or smartphone), a notebook computer, a desktop computer, a tablet computer or device, a remote network server, or so forth. In accordance with another disclosed aspect, a computer program comprises program code means for causing a digital processing device to perform a method according to the invention as described herein. The digital processing device may be a handheld device (e.g., a personal data assistant or smartphone), a notebook computer, a desktop computer, a tablet computer or device, a remote network server, or so forth.
One advantage resides in a clinical decision support (CDS) system providing clinical recommendations based on probabilistic analysis of one or more cellular signaling pathway(s), for example using a Bayesian network model of a Wnt pathway, an ER pathway, an AR pathway and/or a Hedgehog pathway.
Another advantage resides in improved assessment of cellular signaling pathway activity that is less susceptible to error.
Another advantage resides in providing a CDS system recommending targeted treatment for loss of regulation of a cellular signaling pathway.
Another advantage resides in providing a CDS system that is designed to detect loss of regulation for a particular cellular signaling pathway, such as a Wnt pathway, an ER pathway, an AR pathway or a Hedgehog pathway, and is readily adapted to provide recommendations for different types of cancer sourced by that particular cellular signaling pathway.
The present invention as described herein can, e.g., also advantageously be used in connection with - diagnosis based on predicted (inferred) activity;
- prognosis based on predicted (inferred) activity; - drug prescription based on predicted (inferred) activity;
- prediction of drug efficacy based on predicted (inferred) activity;
- prediction of adverse effects based on predicted (inferred) activity;
- monitoring of drug efficacy; - drug development;
- assay development;
- pathway research;
- cancer staging;
- enrollment of subject in a clinical trial based on predicted (inferred) activity; - selection of subsequent test to be performed, and/or;
- selection of companion diagnostics tests.
Further advantages will be apparent to those of ordinary skill in the art upon reading and understanding the attached figures, the following description and, in particular, upon reading the detailed examples provided herein below.
Figure 1 shows a simple Bayesian network representing part of a cellular signaling pathway. The cellular signaling pathway is symbolized by a transcription factor (TF) complex and the target genes produced as a result of the transcription factor complex being present. The probabilistic relationship between the TF element and a target gene in case of binary discretization can be represented by a conditional probability table as depicted in the diagram. Figure 2 shows an illustrative Bayesian network describing a hypothetical cellular signaling pathway. Both the upstream proteins and downstream target mR A nodes are depicted in the diagram. The upstream proteins serve as input in the transcription factor complex, whereas the target mRNAs are the transcription factor complex's output nodes. shows an illustrative example of a Bayesian network representation of a single cellular signaling pathway with multiple transcription factor complex or a multiple cellular signaling pathways with their own transcription factor complex combined into one Bayesian network or a combination thereof. shows an example of a Bayesian network illustrating a simple representation of a cellular signaling pathway similar to Figure 1. Now additional nodes have been attached to represent the translation of target mRNA into target proteins. shows an illustrative of a Bayesian network illustrating another simple representation of a cellular signaling pathway. The pathway is represented using the transcription factor complex and its target protein levels. shows the illustrative Bayesian network of Figure 1 with an additional layer of nodes representing the probesets on a microarray chip connecting the probes' intensities to the corresponding target mRNA levels. shows an illustrative example of a variant embodiment of the Bayesian network of Figure 1 which includes nodes representing methylation and copy number variations as examples for additional information nodes for in this particular example any of the included target mRNA levels. shows a predicted Wnt pathway activity of the Bayesian network and nearest centroid method as described herein in a data set of colon samples (GSE20916).
Figure 9 shows a predicted Wnt pathway activity of the Bayesian network and
nearest centroid method as described herein in a data set of colon samples (GSE4183). Figure 10 shows a predicted Wnt pathway activity of the Bayesian network and nearest centroid method as described herein in a data set of colon samples (GSE15960). Figure 11 shows a predicted Wnt pathway activity of the Bayesian network and
nearest centroid method as described herein in a data set of breast cancer samples (GSE 12777).
Figure 12 shows a predicted Wnt pathway activity of the Bayesian network and
nearest centroid method as described herein in a data set of breast cancer samples (GSE21653).
Figure 13 shows a predicted Wnt pathway activity of the Bayesian network and
nearest centroid method as described herein in a data set of liver cancer samples (GSE9843).
Figure 14 shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of colon samples (GSE20916).
Figure 15 shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of colon samples (GSE4183).
Figure 16 shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of colon samples (GSE15960). shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of breast cancer samples (GSE12777). shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of liver cancer samples (GSE9843). shows a predicted Wnt pathway activity using a Bayesian network using the target genes of the evidence curated list compared to the target genes of the broad literature list as described herein in a data set of medulloblastoma samples (GSE 10327). diagrammatically shows a clinical decision support (CDS) system configured to assess one or more cellular signaling pathway(s) as disclosed herein (exemplary shown for Wnt pathway). shows a predicted Wnt pathway activity in colon samples from GSE4183. shows a predicted Wnt pathway activity in medulloblastoma samples from GSE10327. shows a predicted Wnt pathway activity in liver cancer samples from GSE9843. shows a predicted Wnt pathway activity in breast cancer cell lines from GSE12777. shows a predicted ER pathway activity in breast cancer cell lines from GSE12777. shows a predicted ER pathway activity in breast cancer samples from GSE12276. shows a predicted ER pathway activity in cancer cell lines from GSE36133. shows a predicted Hedgehog pathway activity in cancer cell lines from GSE34211. shows a predicted Hedgehog pathway activity in medulloblastoma samples from GSE10327. shows a predicted Hedgehog pathway activity in breast cancer samples from GSE12276. shows a predicted ER pathway activity in MCF7 and Tamoxifen resistant cell lines from GSE21618. shows a predicted ER pathway activity in a time series of estrogen- stimulated MCF7 cell line samples from GSE11324. shows Wnt, ER and Hedgehog pathway activity in luminal A samples of GSE12276. shows Wnt, ER and Hedgehog pathway activity in basal samples of GSE12276. shows a predicted Wnt pathway activity in colon samples from GSE20916. shows a predicted ER pathway activity in MCF7 cell lines stimulated with estrogen (E2) or a negative control (EtOH) (GSE9253). Figure 37 shows Kaplan-Meier survival curves of patients from the GSE12276 data set grouped according to pathway activity.
Figure 38 shows a predicted AR pathway activity in LNCaP cell lines treated with different treatment regimes from GSE7708.
Figure 39 shows a predicted AR pathway activity in prostate cancer samples from
GSE17951.
Figure 40 shows a predicted AR pathway activity in breast cancer samples from
GSE12276.
Figure 41 shows a predicted AR pathway activity in the GSE36133 data set containing cell line samples representing various cancer types.
Figure 42 shows a predicted AR pathway activity in the GSE34211 data set containing cell line samples representing various cancer types.
The following examples merely illustrate particularly preferred methods and selected aspects in connection therewith. The teaching provided therein may be used for constructing several tests and/or kits, e.g. to detect, predict and/or diagnose the abnormal activity of one or more cellular signaling pathways. Furthermore, upon using methods as described herein drug prescription can advantageously be guided, drug prediction and monitoring of drug efficacy (and/or adverse effects) can be made, drug resistance can be predicted and monitored, e.g. to select subsequent test(s) to be performed (like a companion diagnostic test). The following examples are not to be construed as limiting the scope of the present invention.
Example 1: Bayesian network construction
As disclosed herein, by constructing a probabilistic model (e.g., the illustrative Bayesian model shown in Figure 6) and incorporating conditional probabilistic relationships between expression levels of a number of different target genes and the activity of the cellular signaling pathway, such a model can be used to determine the activity of the cellular signaling pathway with a high degree of accuracy. Moreover, the probabilistic model can be readily updated to incorporate additional knowledge obtained by later clinical studies, by adjusting the conditional probabilities and/or adding new nodes to the model to represent additional information sources. In this way, the probabilistic model can be updated as appropriate to embody the most recent medical knowledge.
One of the simplest Bayesian network models for representing a cellular signaling pathway would be a two level model including the transcription factor element and the associated target genes (see Figure 1). The transcription factor complex element is a representation of the level of the transcription factor complex. The protein level of the transcription factor element is connected to a number of mR A levels of the transcription factor's target genes (in this exemplary Bayesian network only three target genes are depicted, that are known to be expressed in the tissue in case the transcription factor is available). It is to be understood that many, most, or all of the pathway's target genes (in case of the Wnt, ER, Hedgehog and AR pathways particularly the target genes mentioned in Table 1, Table 2, Table 3 and Table 4 respectively) are analogously regulated by the TF element. The relationships between the level of the TF element and the mRNA levels of the target genes are modeled in the Bayesian network by the edges. For each of the target genes, a conditional probabilistic distribution specifies how the gene's mRNA level depends on the level of the TF element.
The levels of the TF element and target genes may be variously represented. One option is to use a binary discretization, into states "absent" and "present" for the TF element, and "down" and "up" for a target gene's mRNA level (see Figure 1). The probabilistic relationship between the TF element and a target gene can then be represented by a conditional probability table (as indicated in the same figure). Instead of a binary discretization, levels can also represented as continuous level values, or as quantized values having three or more quantization levels (e.g. "down", "normal" and "up" for target genes).
The foregoing illustration of a simple Bayesian network is just an illustrative embodiment of the Bayesian network model (Figure 1). In general, a Bayesian network model comprises a directed, acyclic graph comprising nodes connected by edges. Each node represents an information item pertaining to the pathway at hand (or, more generally, to the cellular signaling pathway). Pathway element nodes each represent a genomic or proteomic element of the cellular signaling pathway. By way of illustrative example, a pathway element node may represent one of but not restricted to: a protein, a protein complex, an mR A molecule transcribed from a target gene of the cellular signaling pathway, a methylated gene, a phosphorylated protein, a phosphorylated protein complex, or so forth. As discussed later herein, several other types of nodes, but not limited to the examples given, may be included in the Bayesian network to represent other types of information such as specific measurement datum elements, gene variation occurrences, or so forth.
Additional "upstream" levels representing regulatory proteins (in active or inactive state) of the pathway are typically added if knowledge of the level of such a protein could be probative for determining the clinical decision support recommendation. For example, the inclusion of the proteins elementary to the transcription factor or essential proteins upstream of the transcription factor in the Bayesian network (see Figure 2) could be useful if a drug is available that specifically targets such proteins, rather than the pathway as a whole. The transcription factor (TF) element is believed to be a protein complex (that is, a combination of proteins bound together in a specific structure that performs the function of regulating transcription from the target genes) in the majority of the signaling pathways. For other pathways, the TF element may be a single protein. In addition, signaling pathways may exert their activity through more than one transcription factor resulting in a more complex Bayesian network with multiple transcription factors feeding into the target gene(s) (see Figure 3 for a hypothetical illustration of multiple transcription factor elements influencing target gene transcription). Such a multi- transcription factor Bayesian network may also be the result of a combination of pathways combined in one Bayesian network.
Additional information nodes further downstream of the target genes may be included in the Bayesian network as well. An illustrative example of this is the translation of target gene's mRNA into proteins (Figure 4) or protein level nodes of the target gene as surrogate node of the target gene's mRNA level (Figure 5). The mRNA molecules of the target gene are translated by interaction with ribosome molecules to form proteins corresponding to the mRNA molecules and corresponding to the target genes. This is expression of the target genes at the protein level. Measurement of the protein level by, but not limited to, for example mass-spectrometry, immunohistochemistry, gel electrophoresis techniques may act as evidence for these target protein levels.
The expression level of a target gene may be computed based on the measured intensity of corresponding probesets of a microarray, for example by averaging or by other means of other techniques (e.g. R A sequencing). In some embodiments this computation is integrated into the Bayesian network, by extending the Bayesian network with a node for each probeset that is used and including an edge running to each of these "measurement" nodes from the corresponding target gene node, as described herein with reference to Figure 6.
The probabilistic model may optionally also incorporate additional genomic information, such as information on mutations, copy number variations, gene expression, methylation, translocation information, or so forth, which change genomic sequences which are related to the signaling cascade of the pathway to infer the pathway activity and to locate the defect in the Wnt pathway which causes the aberrant functioning (either activation or inactivity), as described by illustrative reference to Figure 7 for the illustrative case of methylation and copy number data. However, it is to be understood that other types of information regarding the target gene are analogously translated into information nodes. Such genomic information can be available through, but not limited to, RNA sequencing and SNP analysis.
Moreover, it is to be understood that while examples as described later herein pertain to the Wnt, ER, AR and Hedgehog pathway are provided as illustrative examples, the approaches for cellular signaling pathway analysis disclosed herein are readily applied to other cellular signaling pathways besides these pathways, such as to intracellular signaling pathways with receptors in the cell membrane (e.g., the Notch, HER2/PI3K, TGFbeta, EGF, VEGF, and TNF-NFkappaB cellular signaling pathways) and intracellular signaling pathways with receptors inside the cell (e.g., progesterone, retinoic acid, and vitamin D cellular signaling pathways).
Example 2: Comparison of machine learning methods
Here the performance of two types of machine learning techniques are compared to each other with the Wnt pathway taken as an example case: the prediction of Wnt activity by means of a nearest centroid method is compared to the method of choice according to the present invention, which e.g. uses a Bayesian network.
As discussed above the Bayesian network approach was selected based on its advantages residing in the probabilistic approach being able to incorporate the available information in either "soft", e.g. percentages of study subjects exhibiting probative characteristics, and "hard" form, using conditional probabilistic relationships. In addition, the probabilistic model also enables information to be incorporated based on partial (rather than comprehensive) knowledge of the underlying cellular signaling pathway, again through the use of conditional probability tables.
Here it is demonstrated that the inventors added value in the way they included known biological properties and the availability of soft evidence using a Bayesian network compared to other machine learning methods, e.g. nearest centroid classification, a well-known method. Nearest centroid classification is a machine learning method where for each class of training samples an average profile (= centroid) is computed, and next, for a sample to be classified, the label is predicted based on the centroid that is closest (the closest centroid's label is then the prediction result). The two centroids are calculated on the same list of probesets used in the Bayesian network, and for the 'Wnt on' and 'Wnt off centroid they are based on the adenoma samples and the normal colon samples, respectively, of the same fRMA processed data of GSE8671. The log2 -ratio of the two Euclidean distances between a sample and the two centroids was subsequently used to classify samples from various data sets to infer the classification of the samples. This means that a log2 -ratio of 0 corresponds to an equal distance of the sample to the two centroids, a value > 0 corresponds to a sample classified as active Wnt signaling whereas a value < 0 corresponds to a sample identified as having an inactive Wnt signaling pathway.
The Bayesian network was constructed similar to Figure 6 and the procedure described herein. Similarly to this description of the Wnt Bayesian network, the conditional probability tables of the edges between probesets and their respective genes were trained using fRMA processed data of 32 normal colon samples and 32 adenoma samples from data set GSE8671 from the Gene Expression Omnibus (accessible at http://www.ncbi.nlm.nih.gov/geo/, last accessed July 13, 2011). The trained Bayesian network was then tested on various data sets to infer the probability P(Wnt On) that the Wnt pathway is "on", that is, active, which is taken equal to the inferred probability that the Wnt pathway transcription complex is "present".
The trained Bayesian network and nearest centroid model were then tested on various fRMA processed microarray data sets to infer the probability that the Wnt pathway is "on", measured by P(Wnt On) and log2 -ratio of the distances. Summaries of the results of the Bayesian network and the nearest centroid model are shown in Figures 8 to 13. The reader should note that the output metrics of the two methods are not a one-to-one relation, however the sign and the relative magnitude of the output metrics within a method are comparable.
The vast majority of the colon (cancer) samples (GSE20916, GSE4183) are classified equally between the active and inactive Wnt pathway, except for GSE 15960 that had a high fraction of wrongfully classified negative samples in the nearest centroid method (false negatives). This perception of a higher fraction of false negatives is maintained in the other cancer types as well. This is especially true for breast cancer samples (GSE12777, GSE21653) and liver cancer (GSE9843); except for a few exceptions all samples are predicted to have an inactive Wnt pathway which is known to be incorrect in case of basal-type breast cancer and the CTNNB1 liver cancer samples. In some cases, evident in for example GSE15960, the classification could be corrected by lowering and increasing the threshold of the nearest centroid classification. The idea behind this would be that the threshold of Wnt activity might be altered in different tissue-types. However, this would involve additional training of the nearest centroid method to be applicable to other tissue types. One of the strengths of the Bayesian network model is that this tissue- specific training is not required as it is established to be nonspecific regarding tissue-type. Example 3: Selection of target genes
A transcription factor (TF) is a protein complex (that is, a combination of proteins bound together in a specific structure) or a protein that is able to regulate transcription from target genes by binding to specific DNA sequences, thereby controlling the transcription of genetic information from DNA to mRNA. The mRNA directly produced due to this action of the transcription complex is herein referred to as a "direct target gene". Pathway activation may also result in more secondary gene transcription, referred to as "indirect target genes". In the following, Bayesian network models (as exemplary probabilistic models) comprising or consisting of direct target genes, as direct links between pathway activity and mRNA level, are preferred, however the distinction between direct and indirect target genes is not always evident. Here a method to select direct target genes using a scoring function based on available literature data is presented. Nonetheless, accidently selection of indirect target genes cannot be ruled out due to limited information and biological variations and uncertainties
Specific pathway mRNA target genes were selected from the scientific literature, by using a ranking system in which scientific evidence for a specific target gene was given a rating, depending on the type of scientific experiments in which the evidence was accumulated. While some experimental evidence is merely suggestive of a gene being a target gene, like for example a mRNA increasing on an microarray of an embryo in which it is known that the Hedgehog pathway is active, other evidence can be very strong, like the combination of an identified pathway transcription factor binding site and retrieval of this site in a chromatin immunoprecipitation (ChIP) assay after stimulation of the specific pathway in the cell and increase in mRNA after specific stimulation of the pathway in a cell line.
Several types of experiments to find specific pathway target genes can be identified in the scientific literature:
1. ChIP experiments in which direct binding of a pathway-transcription factor to its binding site on the genome is shown. Example: By using chromatin- immunoprecipitation (ChIP) technology subsequently putative functional TCF4 transcription factor binding sites in the DNA of colon cell lines with and without active Wnt pathway were identified, as a subset of the binding sites recognized purely based on nucleotide sequence. Putative functionality was identified as ChlP- derived evidence that the transcription factor was found to bind to the DNA binding site.
2. Electrophoretic Mobility Shift (EMSA) assays which show in vitro binding of a transcription factor to a fragment of DNA containing the binding sequence.
Compared to ChlP-based evidence EMSA-based evidence is less strong, since it cannot be translated to the in vivo situation. 3. Stimulation of the pathway and measuring mRNA profiles on a microarray or using RNA sequencing, using pathway-inducible cell lines and measuring mRNA profiles measured several time points after induction - in the presence of cycloheximide, which inhibits translation to protein, thus the induced mRNAs are assumed to be direct target genes.
4. Similar to 3, but using quantitative PCR to measure the amounts of mRNAs.
5. Identification of transcription factor binding sites in the genome using a
bioinformatics approach. Example for the Wnt pathway: Using the known TCF4- beta catenin transcription factor DNA binding sequence, a software program was run on the human genome sequence, and potential binding sites were identified, both in gene promoter regions and in other genomic regions.
6. Similar as 3, only in the absence of cycloheximide.
7. Similar to 4, only in the absence of cycloheximide.
8. mRNA expression profiling of specific tissue or cell samples of which it is known that the pathway is active, however in absence of the proper negative control condition.
In the simplest form one can give every potential target mRNA 1 point for each of these experimental approaches in which the target mRNA was identified.
Alternatively, points can be given incrementally, meaning one technology 1 point, second technology adds a second point, and so on. Using this relatively ranking strategy, one can make a list of most reliable target genes.
Alternatively, ranking in another way can be used to identify the target genes that are most likely to be direct target genes, by giving a higher number of points to the technology that provides most evidence for an in vivo direct target gene, in the list above this would mean 8 points for experimental approach 1), 7 to 2), and going down to one point for experimental approach 8. Such a list may be called "general target gene list".
Despite the biological variations and uncertainties, the inventors assumed that the direct target genes are the most likely to be induced in a tissue-independent manner. A list of these target genes may be called "evidence curated target gene list". These curated target lists have been used to construct computational models that can be applied to samples coming from different tissue sources.
The "general target gene list" probably contains genes that are more tissue specific, and can be potentially used to optimize and increase sensitivity and specificity of the model for application at samples from a specific tissue, like breast cancer samples.
The following will illustrate exemplary how the selection of an evidence curated target gene list specifically was constructed for the ER pathway.
For the purpose of selecting ER target genes used as input for the "model", the following three criteria were used:
1. Gene promoter/enhancer region contains an estrogen response element (ERE) motif:
a. The ERE motif should be proven to respond to estrogen, e.g., by means of a transient transfection assay in which the specific ERE motif is linked to a reporter gene, and
b. The presence of the ERE motif should be confirmed by, e.g., an enriched motif analysis of the gene promoter/enhancer region.
ER (differentially) binds in vivo to the promoter/enhancer region of the gene in question, demonstrated by, e.g., a ChlP/CHIP experiment or a chromatin immunoprecipitation assay:
a. ER is proven to bind to the promoter/enhancer region of the gene when the ER pathway is active, and
b. (preferably) does not bind (or weakly binds) to the gene promoter/enhancer region of the gene if the ER pathway is not active. The gene is differentially transcribed when the ER pathway is active, demonstrated by, e.g.,
a. fold enrichment of the mRNA of the gene in question through real time PCR, or microarray experiment, or
b. the demonstration that RNA Pol II binds to the promoter region of the gene through an immunoprecipitation assay.
The selection was done by defining as ER target genes the genes for which enough and well documented experimental evidence was gathered proving that all three criteria mentioned above were met. A suitable experiment for collecting evidence of ER differential binding is to compare the results of, e.g., a ChlP/CHIP experiment in a cancer cell line that responds to estrogen (e.g., the MCF-7 cell line), when exposed or not exposed to estrogen. The same holds for collecting evidence of mRNA transcription.
The foregoing discusses the generic approach and a more specific example of the target gene selection procedure that has been employed to select a number of target genes based upon the evidence found using above mentioned approach. The lists of target genes used in the Bayesian network models for exemplary pathways, namely the Wnt, ER, Hedgehog and AR pathways are shown in Table 1, Table 2, Table 3 and Table 4, respectively.
The target genes of the ER pathway used for the Bayesian network model of the ER pathway described herein (shown in Table 2) contain a selection of target genes based on their literature evidence score; only the target genes with the highest evidence scores (preferred target genes according to the invention) were added to this short list. The full list of ER target genes, including also those genes with a lower evidence score, is shown in Table 5.
A further subselection or ranking of the target genes of the Wnt, ER, Hedgehog and AR pathways shown in Table 1, Table 2, Table 3 and Table 4 was performed based on a combination of the literature evidence score and the odds ratios calculated using the trained conditional probability tables linking the probeset nodes to the corresponding target gene nodes. The odds ratio is an assessment of the importance of the target gene in inferring activity of the pathways. In general, it is expected that the expression level of a target gene with a higher odds ratio is likely to be more informative as to the overall activity of the pathway as compared with target genes with lower odds ratios. However, because of the complexity of cellular signaling pathways it is to be understood that more complex interrelationships may exist between the target genes and the pathway activity - for example, considering expression levels of various combinations of target genes with low odds ratios may be more probative than considering target genes with higher odds ratios in isolation. In Wnt, ER, Hedgehog and AR modeling reported herein, it has been found that the target genes shown in Table 6, Table 7, Table 8 and Table 9 are of a higher probative nature for predicting the Wnt, ER, Hedgehog and AR pathway activities as compared with the lower-ranked target genes (thus, the target genes shown in Tables 6 to 9 are particularly preferred according to the present invention). Nonetheless, given the relative ease with which acquisition technology such as microarrays can acquire expression levels for large sets of genes, it is contemplated to utilize some or all of the target genes of Table 6, Table 7, Table 8 and Table 9, and to optionally additionally use one, two, some, or all of the additional target genes of ranks shown in Table 1, Table 2, Table 3 and Table 4, in the Bayesian model as depicted in Figure 6.
Table 1. Evidence curated list of target genes of the Wnt pathway used in the
Bayesian network and associated probesets used to measure the mRNA expression level of the target genes (# = sequence number in accompanying sequence listing).
Evidence curated list of target genes of the ER pathway used in the Bayesian network and associated probesets used to measure the mRNA expression level of the target genes (# = sequence number in accompanying sequence listing). Evidence curated list of target genes of the Hedgehog pathway used in the Bayesian network and associated probesets used to measure the mRNA expression level of the target genes (# = sequence number in accompanying sequence listing).
Evidence curated list of target genes of the AR pathway used in the Bayesian network and associated probesets used to measure the mRNA expression level of the target genes (# = sequence number in accompanying sequence listing).
Gene symbols of the ER target genes found to have significant literature evidence (= ER target genes longlist) (# = sequence number in accompanying sequence listing).
Table 6. Shortlist of Wnt target genes based on literature evidence score and odds ratio (# = sequence number in accompanying sequence listing).
Table 7. Shortlist of ER target genes based on literature evidence score and odds ratio (# = sequence number in accompanying sequence listing).
Shortlist of Hedgehog target genes based on literature evidence score odds ratio (# = sequence number in accompanying sequence listing).
Table 9. Shortlist of AR target genes based on literature evidence score and odds ratio (# = sequence number in accompanying sequence listing).
Example 4: Comparison of evidence curated list and broad literature list
The list of Wnt target genes constructed based on literature evidence following the procedure described herein (Table 1) is compared to another list of target genes not following above mentioned procedure. The alternative list is a compilation of genes indicated by a variety of data from various experimental approaches to be a Wnt target gene published in three public sources by renowned labs, known for their expertise in the area of molecular biology and the Wnt pathway. The alternative list is a combination of the genes mentioned in table S3 from Hatzis et al. (Hatzis P, 2008), the text and table SIA from de Sousa e Melo (de Sousa E Melo F, 2011) and the list of target genes collected and maintained by Roel Nusse, a pioneer in the field of Wnt signaling (Nusse, 2012). The combination of these three sources resulted in a list of 124 genes (= broad literature list, see Table 10). Here the question whether the performance in predicting Wnt activity in clinical samples by the algorithm derived from this alternative list is performing similarly or better compared to the model constructed on the basis of the existing list of genes (= evidence curated list, Table 1) is discussed.
Table 10. Alternative list of Wnt target genes (= broad literature list) (# = sequence number in accompanying sequence listing).
Target Reference # Target gene Reference #
ADH6 de Sousa e Meio et al. L1CAM Nusse 125
LB H
ADRA2 c Hatzis et al. 4 Nusse 126
The next step consisted of finding the probesets of the Affymetrix® GeneChip Human Genome U133 Plus 2.0 array that corresponds with the genes. This process was performed using the Bioconductor plugin in R and manual curation for the probesets relevance based on the UCSC genome browser, thereby removing e.g. probesets on opposite strands or outside gene exon regions. For two of the 124 genes there are no probesets available on this microarray-chip and therefore could not be inserted in the Bayesian network, these are LOC283859 and WNT3A. In total 287 probesets were found to correspond to the remaining 122 genes (Table 1 1).
Table 11. Probesets associated with the Wnt target genes in the broad literature gene list (# = sequence number in accompanying sequence listing).
Subsequently the Bayesian network was constructed similar to Figure 6 and the procedure explained herein. Similarly to the description of the Wnt Bayesian network based on the evidence curated list, the conditional probability tables of the edges between probesets and their respective genes, both the evidence curated list and the broad literature list, were trained using fRMA processed data of 32 normal colon samples and 32 adenoma samples from data set GSE8671 from the Gene Expression Omnibus (accessible at http://www.ncbi.nlm.nih.gov/geo/, last accessed July 13, 2011).
The trained Bayesian networks were then tested on various data sets to infer the probability P(Wnt On) that the Wnt pathway is "on", i.e., active, which is taken equal to the inferred probability that the Wnt pathway transcription complex is "present".
Summarized results of the trained broad literature model and the evidence curated model are shown in Figures 14-19.
Evidently, it could be deduced that the broad literature model generally predicts more extreme probabilities for Wnt signaling being on or off. In addition, the alternative model predicts similar results for the colon cancer data sets (GSE20916, GSE4183, GSE15960), but more than expected samples with predicted active Wnt signaling in breast cancer (GSE 12777), liver cancer (GSE9843) and medulloblastoma sample (GSE 10327) data sets.
In conclusion, the broad literature target genes list results in approximately equally well predictions of Wnt activity in colon cancer on the one hand, but worse predictions (too many false positives) in other cancer types on the other hand. This might be a result of the alternative list of targets genes being too much biased towards colon cells specifically, thus too tissue specific; both de Sousa E Melo et al. and Hatzis et al. main interest was colorectal cancer although non-colon- specific Wnt target genes may be included. In addition, non-Wnt-specific target genes possibly included in these lists may be a source of the worsened predictions of Wnt activity in other cancer types. The alternative list is likely to contain more indirectly regulated target genes, which probably makes it more tissue specific. The original list is tuned towards containing direct target genes, which are most likely to represent genes that are Wnt sensitive in all tissues, thus reducing tissue specificity.
Example 5: Training and using the Bayesian network
Before the Bayesian network can be used to infer pathway activity in a test sample, the parameters describing the probabilistic relationships between the network elements have to be determined. Furthermore, in case of discrete states of the input measurements, thresholds have to be set that describe how to do the discretization.
Typically, Bayesian networks are trained using a representative set of training samples, of which preferably all states of all network nodes are known. However, it is impractical to obtain training samples from many different kinds of cancers, of which it is known what the activation status is of the pathway to be modeled. As a result, available training sets consist of a limited number of samples, typically from one type of cancer only. To allow the Bayesian network to generalize well to other types of samples, one therefore has to pay special attention to the way the parameters are determined, which is preferably done as follows in the approach described herein.
For the TF node, the (unconditional) probability of being in state "absent" and "present" is given by the expected occurrence on a large set of samples. Alternatively, one can set them to 0.5, as is done in Figure 1, in order to have no bias for a positive or negative outcome.
For the target gene nodes, the conditional probabilities are set as in Figure 1. If the TF element is "absent", it is most likely that the target gene is "down", hence a probability of 0.95 is chosen for this, and a probability of 0.05 for the target gene being "up". The latter (non-zero) probability is to account for the (rare) possibility that the target gene is regulated by other factors or accidentally observed "up" (e.g. because of measurement noise). If the TF element is "present", then with a fair probability of 0.70 the target gene is "up", and with a probability of 0.30 the target gene is "down". The latter values are chosen this way, because there can be several reasons why a target gene is not highly expressed even though the TF element is present, for instance because the gene's promoter region is methylated. In the case that a target gene is not up-regulated by the TF element, but down-regulated, the probabilities are chosen in a similar way, but reflecting the down-regulation upon presence of the TF element.
For the Bayesian network model as given in Figure 6, where the intensities of the probesets form the input measurements, one finally has to determine the parameters for discretization and for the conditional probability tables relating the probesets' intensities to the mR A levels of the respective target genes. Both of these are based on training data in the current invention. For the discretization of a probeset' s intensity level into states "low" and "high", a suitable threshold is determined that best separates the intensity values in a set of training samples where the pathway is activated ("on" samples) from the intensity values in a set of training samples in which it is not ("off samples). Finally, the conditional probability tables describing the probabilities of a probeset to have a "low" or "high" intensity depending on the "down" or "up" state of the respective target gene is done by counting the number of "on" and "off samples with an intensity value of the probeset below and above the respective threshold. This is known in the literature as the frequentist approach. A dummy count is added to each group to prevent entries in the conditional probability tables with a value of zero, to prevent extreme behavior of the Bayesian network.
After the Bayesian network has been trained, it can be applied on a test sample as follows, considering the Bayesian network of Figure 6, and assuming the microarray measurements relating to the probesets are available. The first step is to discretize the input measurements, by comparing each probeset's intensity in the test sample to the respective threshold as described above. This comparison may be done in a hard way, setting each probeset to either "low" or "high" intensity (called 'hard evidence'), or it can be done in a soft way, assuming some uncertainty (noise) in the measurement, setting for each probeset a probability of being "low" or "high" (called 'soft evidence'). For instance, the soft evidence of a probeset with an intensity just below the threshold may be a probability of 0.8 of being "low" and a probability of 0.2 of being "high", based on a suitable estimate of the noise and the difference to the threshold.
Next, this hard or soft evidence is supplied to a suitable inference engine for Bayesian networks, for instance based on a junction tree algorithm (see (Neapolitan, 2004)). Such an engine can then infer the updated probability of the TF element being "absent" or "present", given the provided evidence. The inferred probability of the TF element being "present" is then interpreted as the estimated probability that the respective pathway is active.
Preferably, the training of the Bayesian network models of the Wnt, ER, Hedgehog and AR pathways is done using public data available on the Gene Expression Omnibus (accessible at http://www.ncbi.nlm.nih.gov/geo/, cf. above).
The Wnt Bayesian network was exemplary trained using 32 normal colon samples considered to have an inactive Wnt pathway and 32 confirmed adenoma samples known to have an active Wnt pathway (GSE8671 data set).
The Bayesian network model of the ER pathway was exemplary trained using 4 estrogen-deprived MCF7 samples, known to have an inactive ER pathway, and 4 estrogen-stimulated MCF7 samples, regarded to have an active ER pathway, from the GSE8597 data set also accessible at the Gene expression Omnibus.
The Bayesian network model of the Hedgehog pathway was exemplary trained using 15 basal cell carcinoma samples confirmed to have an active Hedgehog pathway and 4 normal skin cells samples representing samples with an inactive Hedgehog pathway available in the GSE7553 data set.
The Bayesian network model of the AR pathway was exemplary trained using 3 samples with positive AR activity, LNCaP cell lines stimulated with
Dihydrotestosterone (DHT), a potent AR pathway activator, and 3 non-stimulated LNCaP cell lines representing the inactive AR pathway case.
With reference to Figure 35 and Figure 36, the trained Bayesian network models of the Wnt and ER pathway were used to predict the pathway activities in similar samples (colon samples and MCF7 breast cancer cell line for the Wnt and ER Bayesian network, respectively) not used in the training procedure as described herein (no appropriate data set for the Hedgehog Bayesian network was found). The predicted pathway activities of the vast majority of the samples should be in line with the clinically expected pathway activities for the model to be validated.
Figure 35 shows the predicted Wnt activities, depicted as the logit of P(Wnt on) on the vertical axis, for the samples, illustrated by the bars on the horizontal axis, of the colon samples grouped by classification, indicated by the bar's color, in the GSE20916 data set. All normal colon samples are rightfully predicted to have an inactive pathway
(score < 0), based on it being a sample of healthy tissue. All but four samples alleged to have an active pathway are predicted to have an active Wnt pathway.
In Figure 36 the validation results of the trained ER Bayesian network model is shown for two microarrays measured using a MCF7 breast cancer cell line sample, one stimulated with estrogen (E2) the other one with a negative control (EtOH), originating from the GSE9253 data set. In agreement with the alleged ER activity, the sample stimulated with estrogen is predicted to have an active ER pathway, whereas the negative control predicts an inactive ER pathway.
Further details and examples for using trained Bayesian networks (e.g. of
Wnt, ER, AR and Hedgehog pathway) to predict the respective pathway activities are explained in Example 6 below.
The above mentioned training process can be employed to other Bayesian networks of clinical applications. Here it is shown and proven to work for the Bayesian network models constructed using herein disclosed method representing cellular signaling pathways, more specifically the Wnt, ER, AR and Hedgehog pathways.
Example 6: Diagnosis of (abnormal) pathway activity
The following will exemplary illustrate how to use e.g. Bayesian network models to diagnose the activity of a cellular signaling pathway.
The Bayesian networks of the Wnt, ER, Hedgehog and AR pathway, constructed using a node for the transcription factor presence, a layer of nodes representing the target genes' mRNA and a layer of nodes representing the probesets' intensities corresponding to the target genes (Table 1, Table 2, Table 3 and Table 4), analogous to Figure 6 described herein, and trained as described herein, were used to predict the pathways activity as "on", that is active, or "off, that is inactive, in various, previously not used for training, data sets to infer how well the inference component operates. The predicted pathway activity scores are correlated with clinical knowledge. Result summaries for a selection of the test runs are shown in Figures 21 seq.
With reference to Figures 21 seq., pathway activity inference results for medical tissue samples using the Bayesian network model described herein are shown.
Figure 21 shows results for Wnt activity tests on colon samples data set
GSE4183. The Bayesian network model yielded high values of P(Wnt On) for the adenoma samples, and low values for normal samples, which corresponds with the
(patho)physiology of adenoma and healthy tissue. Healthy tissue has a slow cell proliferation and thus a low Wnt activity relative to adenomatous tissue which has a rapid cell proliferation and thus high Wnt activity. For the IBD samples, the Bayesian network model showed low Wnt pathway activity (P(Wnt On)~0) for all but one sample. Again, this is consistent with the IBD samples not undergoing rapid cell proliferation. For the colorectal cancer cell samples the results were mixed, with high Wnt pathway activity being detected in about one half of these samples, but this can be a result of other pathways assuming the role of tumor driver when benign adenomatous tissue becomes malignant cancerous tissue, or sample analysis problems e.g. the sample containing too much non- tumor tissue, or the mRNA being partly degraded
The Bayesian network model used in the experiments reported herein was trained using the colon samples data set GSE8671. However, the Wnt pathway is present (albeit possibly inactive) in other cell types. It was therefore considered possible that the Bayesian network might be applicable to infer abnormally high Wnt pathway activity correlative with other types of cancers. The rationale for this is that, although the Bayesian network model was trained using colon samples, it is based on first principles of the operation of the Wnt pathway present (albeit possibly inactive) in other cell types. Figures 22 - 24 show some results investigating such "cross tissue type" inferences.
Figure 22 shows results for tests using the Bayesian network model trained using colon samples being applied to infer Wnt pathway activity in medulloblastoma samples (data set GSE 10327). The samples included in this data set have been further characterized in several subsets, one of them being samples with the Wnt pathway being active. The Bayesian network of the Wnt pathway predicts the group of Wnt-active samples having an active Wnt pathway, whereas the other samples where predicted correctly to have an inactive Wnt pathway. The test results using the Wnt Bayesian network model in a data set containing liver cancer samples (GSE9843) is shown in Figure 23. Here the samples are grouped by the following a priori annotations assigned by the GSE9843 data set:
"CT NB1", "Inflammation", "Polysomy chr7", "Proliferation", and "Unannotated". The samples of the "Inflammation" group are uniformly inferred to not have abnormally high Wnt pathway activity, as expected since the inflammation condition does not entail rapid cell proliferation. Samples labeled "Polysomy chr7" are also uniformly inferred to not have abnormally high Wnt pathway activity. Polysomy of chromosome number 7 means that there are more than two number 7 chromosomes. As there is no reason to expect this polysomy condition to impact the Wnt pathway, it is not unexpected that these samples do not have abnormally high Wnt pathway activity.
About one in five of the samples labeled "Proliferation" have P(Wnt On)>0.5. Proliferation suggests a state of rapid cellular multiplication. Such a state may be associated with abnormally high Wnt pathway activity, but may also be associated with numerous other possible causes of cell proliferation. Accordingly, about one in five of these samples having abnormally high Wnt pathway activity is not an unreasonable result.
About one half of the samples of the "CTNNB1" group are inferred by the Bayesian network to have abnormally high Wnt pathway activity. The CTNNB1 gene encodes the beta-catenin protein, which is a regulatory protein of the Wnt pathway, and activating mutations in this gene cause abnormal Wnt activation. Thus, a correlation between the "CTNNB1" group and high Wnt pathway activity is conform expectation.
Figure 24 depicts the test results of the Wnt Bayesian network model described herein for a set of breast cancer samples. In this case three groups of breast cancer cell lines are tested: one group for which the Wnt pathway is a priori known to be operating at an abnormally high level (Wnt on-group); one group for which the Wnt pathway is a priori known to not be operating at an abnormally high level (Wnt off-group); and another group for which the Wnt pathway activity is not a priori known (Unknown group); in addition there is also one sample that is suspected to have a low level of Wnt activation (Wnt suspicious), although there is a conflicting report in the literature that it may have an active Wnt pathway (but this is a minority report; more papers report an inactive Wnt pathway). As seen in Figure 24, the correlation of the inferences provided by the Bayesian network with the a priori knowledge is strong for the Wnt on- and off-groups. Also the rightmost sample of the graph (Wnt suspicious) shows an inference that corresponds to most reports in the literature stating that the Wnt pathway is off. In the case of unknown-group shown in Figure 24, for which there is no a priori knowledge of the Wnt pathway activity, the Bayesian network infers low activity for the Wnt pathway except for one instance for which P(Wnt On)>0.5; literature shows that this cell line has a high expression of the co-receptor LRP6, which may explain that the Wnt pathway is on.
Figure 25 shows the results for the same data set of breast cancer cell lines but now tested for ER activity using the ER Bayesian network trained using MCF7 breast cancer cell lines as described herein. The samples a priori known to have an active Wnt pathway were predicted to have an inactive ER pathway, which is not surprising since the Wnt pathway is already driving the rapid cell multiplication. ER positive samples, on the other hand are found amongst the Wnt off-samples and the unknown-samples. In view of Figure 24, this is not surprising.
The test results of the predictions of the ER Bayesian network trained on breast cancer cell lines for a set of cancer samples (GSE12276) are shown in Figure 26. The breast cancer samples were subdivided in the well-known classifications: Luminal A (LumA), Luminal B (LumB), Human Epidermal Growth Factor Receptor 2 positive (HER2), and basal breast cancer subtype. Tissue samples in the luminal A and luminal B subtypes are known to express ER. It is also in these subtypes that most samples are predicted to have a high ER pathway activity. On the other hand, the samples that are classified to be of the basal subtype are known to have no or low expression of ER, which nicely correlates with no active ER pathway predicted in the basal group samples. In the HER2 group only three samples have a P(ER on)>0.5, whereas the majority of the samples are predicted to have an inactive ER pathway. This correlates well with the fact that the classification is done on the fact that these samples have an amplified HER2 expression; uncontrolled cell replication is presumably driven via HER2 signaling via other cellular signaling pathways than the ER pathway (see for example the Wnt active breast cancer cell lines in Figure 24 or Hedgehog active breast cancer samples in Figure 30).
The ER Bayesian network model constructed and trained as described herein is used to predict the ER pathway activity in a large panel of cell lines of various cancers, the results are shown in Figure 27. As expected only ER active predicted samples were found in the breast cancer cell lines. All other types of cancer cell lines were predicted to have an inactive ER pathway, which is as expected.
The Bayesian network model constructed and trained for the Hedgehog pathway as described herein is used to predict the activity of the Hedgehog pathway for cell lines of various cancer types in the GSE34211-data set. The Hedgehog activity predictions are shown in Figure 28. The highest fractions of positive predicted Hedgehog activity are found in the central nervous system (CNS), skin, endometrium, and uterus cancer types, which is in good agreement with literature knowledge regarding Hedgehog- dependent cell proliferation in these cell types.
Figure 29 shows the predicted Hedgehog activity of the medulloblastoma samples (GSE 10327) that was already analyzed using the Wnt Bayesian network model as described herein. The medulloblastoma samples have been characterized in subclasses, with one of them having an active Hedgehog signaling pathway (identifier: SHH). All of the samples in the SHH subtype are predicted to have an active Hedgehog signaling.
Further, the medulloblastoma samples in the Wnt-subtype were also predicted to have an active Hedgehog pathway. This is in agreement with clinical evidence showing that often both pathways are active in these tumors. Nevertheless, the Wnt Bayesian network was clearly able to correctly predict Wnt activity only in the Wnt-subtype. Thus the
combination of the Wnt and Hedgehog Bayesian network are able to make a correct classification of these two subtypes.
The predicted Hedgehog activity in the GSE 12276 breast cancer samples, earlier used to predict the ER activity using the ER Bayesian network model, using the Hedgehog Bayesian network model is shown in Figure 30. The Hedgehog pathway is predicted to be active in a fraction of the samples of every subtype. This seems odd, but matched to the ER pathway prediction shown in Figure 26 one can see that Hedgehog activity is only predicted in samples not having an active ER pathway. This is in good agreement with the hypothesis that uncontrolled cell proliferation in (breast) tissue can be driven by different signaling pathways.
In summary, the test results for various cancerous tissue samples and cells presented in Figures 21 - 30 strongly suggest that the Bayesian networks of the Wnt, ER and Hedgehog models trained on tissue/pathway specific samples are applicable to analysis of samples of other types of tissue. This can enable cellular signaling pathway analysis to be applied "cross tissue type". Thus, the CDS system 10 (as described herein) is readily applied to assess pathway activity in a range of tissue types other than the tissue type of the samples used to train the Bayesian network model 40 (see e.g. Figure 20 which shows diagrammatically a clinical decision support (CDS) system configured to assess one or more cellular signaling pathways as disclosed herein (exemplary shown for Wnt pathway)). In cases where the inference components 40, 44, 46, 48 indicate the tissue under analysis exhibits abnormally high Wnt, ER or Hedgehog pathway activity, but no tissue specific drug is available, a general Wnt, ER or Hedgehog pathway suppression drug, or a malfunction specific drug, may be considered by the physician based on the recommendation 28 or the recommendation 26, respectively, as provided by the CDS system 10.
Although the results of Figures 21 - 30 indicate cross tissue type
applicability of the Bayesian network model for the Wnt, ER and Hedgehog pathway, it is expected that for clinical applications the Bayesian network models may optionally be updated or adapted to maximize its applicability to the specific tissue type under analysis (e.g., breast tissue or liver tissue). Such updating or adaptation could, for example, entail adjusting the conditional probabilities based on clinical studies of the tissue type under analysis or enrich the evidence curated target gene list, described herein, with tissue specific target genes of the pathway(s) under investigation. Additionally, nodes might be added or removed to better tune the Bayesian network model to the tissue under analysis. Alternatively, different Bayesian network models may be trained ab initio using different training sets for the different tissue types. Furthermore, the results of Figures 21 - 30 illustrate the ability of the process described herein to develop and train Bayesian network models using evidence curated target gene lists of pathways other than Wnt, ER and Hedgehog to predict and diagnose the pathway's activity.
The test results of the AR Bayesian network model constructed and trained as described herein was exemplary used to predict the AR activity in LNCaP prostate cancer cell lines treated with different treatment regimes (GSE7708) (see Figure 38). As expected LNCaP cells not stimulated with DHT results in a predicted inactive AR pathway, whereas LNCaP stimulated cells were correctly predicted to have an active AR pathway and LNCaP cells treated with Bicalutamide, an anti-androgen drug, to have an inhibited AR pathway. The trained Bayesian network of the AR pathway as described herein was also used to predict the probability the AR pathway is active in prostate cancer samples from the GSE17951 data set (results are shown in Figure 39). The majority of the prostate biopsies and tumors were not unexpectedly predicted to have a higher probability of AR activity compared to the controls samples.
The AR Bayesian network model was also applied to a cross-tissue test, viz. the breast cancer samples included in the GSE 12276 data set. Results for this test are shown in Figure 40. A small fraction of the samples, found in every subgroup, are predicted to have an active pathway, whereas the vast majority of the samples had an inactive AR pathway. Remarkably the highest percentage of samples with an active AR pathway are found in the HER2-subgroup, which is not unexpected as it is known from literature that there is crosstalk between the HER2 and AR pathway and the AR pathway can also be induced by HER2-signaling.
The above mentioned AR Bayesian network model was also used to predict the AR pathway's activity in two sets of cell lines samples of various cancer types
(GSE36133 and GSE34211) as depicted in Figure 41 and Figure 42. As expected, the majority of the cell lines were found to have an inactive AR pathway. The exceptions to this are the prostate cancer samples with several cancer cell line samples expressing AR pathway activity. In Table 12 is shown that all AR pathway activity predictions of the prostate cancer samples are in agreement with the known AR activity.
Table 12. Known and predicted AR activity in prostate cancer cell lines in GSE36133 and GSE34211 data sets.
1 GSM844559 j HPET11 [ NO 0.005602
H PET13 replicate 1
H PET13 replicate 2 NO
1 GSM844562 j HPET5 1 NO } 0.007673
™†LNCAP
1 GSM 1844674 1 PCFPTIZER
1 GSM844675 ! PC3 1 NO ] 6.006163 s I I Good N CI50 WYETH | |
Example 7: Prognosis based on pathway activity
Early developmental pathways, like Wnt and Hedgehog, are thought to play a role in metastasis caused by cancer cells which have reverted to a more stem cell like phenotype, called cancer stem cells. Indeed, sufficient evidence is available for the early developmental pathways, such as Wnt pathway, to play a role in cancer metastasis, enabling metastatic cancer cells to start dividing in the seeding location in another organ or tissue. Metastasis is associated with bad prognosis, thus activity of early developmental pathways, such as the Wnt and Hedgehog pathway, in cancer cells is expected to be predictive for bad prognosis. This is supported by the fact that breast cancer patients, from the GSE 12276 data set, that were identified having an active ER pathway but not having an active Wnt or Hedgehog pathway using the Bayesian network models described herein had a better prognosis than patients identified having either an active Hedgehog or Wnt pathway or both, as illustrated by the Kaplan-Meier plot in Figure 37.
Example 8: Therapy planning, prediction of drug efficacy, prediction of adverse effects and monitoring of drug efficacy
The following exemplary illustrates how to use the probabilistic models, in particular Bayesian network models, for therapy planning, prediction of drug efficacy, monitoring of drug efficacy and related activities.
The Bayesian network model of the ER pathway, constructed using a node for the transcription factor presence, a layer of nodes representing the target genes' mRNA levels (Table 2) and a layer of nodes representing the probesets' intensities corresponding to the target genes (Table 2), analogous to Figure 6 described herein, and trained as described herein, were used to predict the pathway activity. The pathway activity is subsequently demonstrated to be correlated with drug efficacy or monitoring drug efficacy. Result summaries are shown in Figures 31 and 32.
Tamoxifen is a drug currently used for the treatment of ER+ (estrogen receptor positive) breast cancer. It acts as a partial antagonist of the estrogen receptor inhibiting the uncontrolled cell proliferation which is thought to be induced by ER signaling. Unfortunately, not every breast cancer responds to treatment with Tamoxifen, despite the demonstration of the presence of ER protein in cancer cells by routine histopathology analysis of cancer tissue slides. Many studies have been conducted to investigate this so-called Tamoxifen resistance. The publicly available GSE21618 data set is the result of one of such study and contains microarray data of Tamoxifen resistant and wildtype MCF7 cell lines under different treatment regimes. The ER Bayesian network model constructed and trained as described herein is used to analyze the Tamoxifen resistant and MCF7 cell lines under different treatment regimes, the results are depicted in Figure 31.
The control Tamoxifen resistant cell line, indicated by TamR.Ctrl, is predicted to have an inactive ER pathway for every time point after Tamoxifen addition (1, 2, 3, 6, 12, 24, and 48 h). It is not surprising that treatment of the Tamoxifen resistant cell line, that is insensitive to Tamoxifen treatment, with Tamoxifen, indicated by TamR.Tam, is ineffective, which is also illustrated by the predicted inactivity of the ER pathway for this group over the same time points. According to analysis of the Tamoxifen resistant cell line (TamR.Ctrl) the driving force of the uncontrolled cell proliferation is not due to active ER signaling; therefore treating it with an ER antagonist will not inhibit cell proliferation. This illustrates that treatment with Tamoxifen is not recommended in case of a negative predicted ER pathway activity.
On the other hand, the wild type MCF7 cell line, known to be Tamoxifen sensitive, treated with 17beta-estradiol (wtl .E2) slowly reacts to the hormone treatment which is visible in the increasing ER positive activity predictions. Treating such a cell line with aromatase inhibitors that are known to inhibit estrogen production will inhibit the ER pathway which is illustrated by the decreasing ER pathway prediction in time. Supporting this are the ER pathway predictions made based on the microarray data from MCF7 samples treated with estrogen for increasing time in the GSE11324 data set, results shown in Figure 32. The above mentioned illustrates the ability of the probabilistic models, in particular the Bayesian network models, to be used for therapy planning, drug efficacy prediction, and monitoring of drug efficacy. However it is to be understood, the same methodology would also apply to predict and monitor adverse effects.
Example 9: Drug development
Similar to therapy response monitoring, a pathway model can be used in drug development to assess the effectiveness of various putative compounds. For instance, when screening many compounds for a possible effect on a certain pathway in a cancer cell line, the respective pathway model can be used to determine whether the activity of the pathway goes up or down after application of the compound or not. Often, this check is done using only one or a few of putative markers of the pathway's activity, which increases the chance of ineffective monitoring of the treatment effect. Furthermore, in follow-up studies on animal or patient subjects, the pathway models can be used similarly to assess the effectiveness of candidate drugs, and to determine an optimal dose to maximally impact pathway activity.
An example of ineffective monitoring of new drug compounds is illustrated by the predicted AR pathway activity in the GSE7708 samples as shown in Figure 38. In this study two possible drug compounds to inhibit AR pathway activity, denoted by Polyamide 1 and Polyamide 2, have been developed. It has been demonstrated that these two polyamides are capable of inhibition of the AR pathway based on the findings that polyamides bind to the Androgen Response Element (ARE) and inhibit expression of KLK3 (= PSA), a well-known marker for AR activity also included in the target gene selection as described herein, as well as ~ 35 % of the transcripts that were induced by DHT. In contrast the Bayesian network model of the AR pathway predicted these samples to still have an active AR pathway. Investigating the inferred probabilities of the target genes being upregulated using the AR Bayesian network model indicated that KLK3 in contrast to the other target genes was downregulated in accordance to the findings whereas all other target genes (except for AR, GUCY1A3 and TMPRSS2 in case of Polyamide 1) were clearly differentially expressed in the Polyamide 1 and Polyamide 2 treated samples. In other words, only one marker for AR activity, KLK3, was downregulated, whereas the majority of the identified target genes were still upregulated indicating the AR pathway is still largely intact and thus active. By taking into account a larger number of target genes based on literature evidence the inventors were able to show that the inhibition of AR activity of the polyamides is limited and that only KLK3 expression is clearly
downregulated using these polyamides. Moreover, this illustrates the value of a systematic approach using a Bayesian network model compared to a reductionist approach in drug development.
Example 10: Assay development
Instead of applying mentioned Bayesian networks on mRNA input data coming from microarrays or RNA sequencing, it may be beneficial in clinical applications to develop dedicated assays to perform the sample measurements, for instance on an integrated platform using qPCR to determine mRNA levels of target genes. The
RNA/DNA sequences of the disclosed target genes can then be used to determine which primers and probes to select on such a platform.
Validation of such a dedicated assay can be done by using the microarray- based Bayesian networks as a reference model, and verifying whether the developed assay gives similar results on a set of validation samples. Next to a dedicated assay, this can also be done to build and calibrate similar Bayesian network models using niRNA-sequencing data as input measurements.
Example 11: Pathway research and cancer pathophysiology research
The following will illustrate how Bayesian network models can be employed in (clinical) pathway research, that is research interested to find out which pathways are involved in certain diseases, which can be followed up for more detailed research, e.g. to link mutations in signaling proteins to changes in pathway activation (measured with the model). This is relevant to investigate the initiation, growth and evolution and metastasis of specific cancers (the pathophysiology).
The Bayesian network models of the Wnt, ER, Hedgehog and AR pathway, constructed using a node for the transcription factor presence, a layer of nodes representing the target genes' mRNA levels (Table 1, Table 2, Table 3 and Table 4) and a layer of nodes representing the probesets' intensities corresponding to the target genes (Table 1, Table 2, Table 3 and Table 4), analogous to Figure 6 described herein, and trained as described herein, were used to predict the pathway activity of a data set consisting of breast cancer samples (GSE12276).
Suppose the researcher is interested in looking into the cellular signaling pathway or pathways and the specific deregulation(s) that drive(s) the uncontrolled cell proliferation. The researcher can analyze the microarray data using the above mentioned probabilistic models, in particular the Bayesian network models, to find which pathways are presumably the cause of uncontrolled cell proliferation. Shown in Figure 33 and Figure 34 one can see an illustration of such an analysis for the case of Wnt, ER and Hedgehog activity (basal and luminal A samples of the GSE 12276 data set). Subsequently, the researcher can search in more detail to find the exact cause of pathway deregulation.
With reference to Figure 34, the basal samples are known to have triple negative receptor status (ER, PR and HER2), therefore it is not surprising to see that all samples are predicted to have an inactive ER pathway. On the other hand some of the samples are predicted to have either the Wnt or Hedgehog or both active as shown in Figure 34. These predicted pathway activities persuade the researcher to investigate these samples in more detail for e.g. known mutations or other known deregulations in the Wnt and/or Hedgehog pathways.
Another example is given in Figure 33, where the Wnt, ER and Hedgehog activities in the luminal A samples of the GSE 12276 data set are illustrated. Luminal A samples are known to express ER, however this does not necessarily mean the cancerous properties are due to active ER signaling. From the predicted pathway activities one can infer that less than half of the samples have an active ER signaling. However, some of the samples that do not have an active ER signaling are found to have an active Wnt and/or Hedgehog pathway. This might give rise for the researcher to investigate these samples in closer details for defects in the Wnt and/or Hedgehog signaling pathway, respectively.
Some of the samples do not predict any of the included three pathways being active; maybe other pathways are causing the uncontrolled cell proliferations. Also this gives the researcher additional information to search for defects in other pathways.
In summary, the illustrations described herein indicate the ability of trained Bayesian network models (as described above) to support the process of finding the cause of uncontrolled cell proliferation in a more directed method. By employing the Bayesian networks to screen the samples for pathway activities, the predicted pathway activities can pinpoint the possible pathways for the cell proliferation, which can be followed up for more detailed research, e.g. to link mutations in signaling proteins or other known deregulations to changes in activation (as measured with the model).
As described herein, the process to develop and train a Bayesian network of cellular signaling pathways can be used to construct a Bayesian network model for other pathways that could also be employed in connection with the present invention.
Example 12: Enrollment of subject in a clinical trial based on predicted activity
If a candidate drug is developed to, for instance, block the activity of a certain pathway that drives tumor growth, and this drug is going into clinical trial, then a proper selection of the subjects to enroll in such a trial is essential to prove potential effectiveness of the drug. In such a case, patients that do not have the respective pathway activated in their tumors should be excluded from the trial, as it is obvious that the drug cannot be effective if the pathway is not activated in the first place. Hence, a pathway model that can predict pathway activity can be used as a selection tool, to only select those patients that are predicted to have the respective pathway activated.
Example 13: Selection of subsequent test(s) to be performed
If a tumor is analyzed using different pathway models, and the models predict deregulation of a certain pathway, then this may guide the selection of subsequent tests to be performed. For instance, one may run a proximity ligation assay (PLA) to confirm the presence of the respective transcription complex (Soderberg O, 2006). Such a PLA can be designed to give a positive result if two key proteins in a TF complex have indeed bound together, for instance beta-catenin and TCF4 in the TF complex of the Wnt pathway.
Another example is that the pathway predicted to be deregulated is analyzed in more detail with respect to the signaling cascade. For instance, one may analyze key proteins in this pathway to determine whether there are mutations in the DNA regions encoding for their respective genes, or one may test for the abundance of these proteins to see whether they are higher or lower than normal. Such tests may indicate what the root cause is behind the deregulation of the pathway, and give insights on which available drugs could be used to reduce activity of the pathway. These tests are selected to confirm the activity of the pathway as identified using the Bayesian model. However selection of companion diagnostic tests is also possible. After identification of the pathway using the model, for targeted therapy choice only those companion diagnostics tests need to be performed (the selection), which are applicable to the identified pathway.
Example 14: Selection of companion diagnostics tests
Similar to the previous example, if a tumor is analyzed and the pathway models predict deregulation of a certain pathway, and optionally a number of additional tests have been performed to investigate the cause of deregulation, then an oncologist may select a number of candidate drugs to treat the patient. However, treatment with such a drug may require a companion diagnostic test to be executed first, for instance to comply with clinical guidelines or to ensure reimbursement of the treatment costs, or because regulatory (FDA) it is required to perform the companion diagnostic test prior to giving the drug. An example of such a companion diagnostic test is the Her2 test for treatment of breast cancer patients with the drug Herceptin (Trastuzumab). Hence, the outcome of the pathway models can be used to select the candidate drugs and the respective companion diagnostic tests to be performed. Example 15: CDS application
With reference to Figure 20 (diagrammatically showing a clinical decision support (CDS) system configured to assess one or more cellular signaling pathways as disclosed herein (exemplary shown for Wnt pathway)), a clinical decision support (CDS) system 10 is implemented as a suitably configured computer 12. The computer 12 may be configured to operate as the CDS system 10 by executing suitable software, firmware, or other instructions stored on a non-transitory storage medium (not shown) such as a hard drive or other magnetic storage medium, an optical disk or other optical storage medium, a random access memory (RAM), read-only memory (ROM), flash memory, or other electronic storage medium, a network server, or so forth. While the illustrative CDS system 10 is embodied by the illustrative computer 12, more generally the CDS system may be embodied by a digital processing device or an apparatus comprising a digital processor configured to perform clinical decision support methods as set forth herein. For example, the digital processing device may be a handheld device (e.g., a personal data assistant or smartphone running a CDS application), a notebook computer, a desktop computer, a tablet computer or device, a remote network server, or so forth. The computer 12 or other digital processing device typically includes or is operatively connected with a display device 14 via which information including clinical decision support recommendations are displayed to medical personnel. The computer 12 or other digital processing device typically also includes or is operatively connected with one or more user input devices, such as an illustrative keyboard 16, or a mouse, trackball, trackpad, touch-sensitive screen (possibly integrated with the display device 14), or other pointer-based user input device, via which medical personnel can input information such as operational commands for controlling the CDS system 10, data for use by the CDS system 10, or so forth.
The CDS system 10 receives as input information pertaining to a medical subject (e.g., a hospital patient, or an outpatient being treated by an oncologist, physician, or other medical personnel, or a person undergoing cancer screening or some other medical diagnosis who is known or suspected to have a certain type of cancer such as colon cancer, breast cancer, or liver cancer, or so forth). The CDS system 10 applies various data analysis algorithms to this input information in order to generate clinical decision support recommendations that are presented to medical personnel via the display device 14 (or via a voice synthesizer or other device providing human-perceptible output). In some embodiments, these algorithms may include applying a clinical guideline to the patient. A clinical guideline is a stored set of standard or "canonical" treatment recommendations, typically constructed based on recommendations of a panel of medical experts and optionally formatted in the form of a clinical "flowchart" to facilitate navigating through the clinical guideline. In various embodiments the data processing algorithms of the CDS 10 may additionally or alternatively include various diagnostic or clinical test algorithms that are performed on input information to extract clinical decision recommendations, such as machine learning methods disclosed herein.
In the illustrative CDS systems disclosed herein (e.g., CDS system 10), the CDS data analysis algorithms include one or more diagnostic or clinical test algorithms that are performed on input genomic and/or proteomic information acquired by one or more medical laboratories 18. These laboratories may be variously located "on-site", that is, at the hospital or other location where the medical subject is undergoing medical examination and/or treatment, or "off-site", e.g. a specialized and centralized laboratory that receives (via mail or another delivery service) a sample of tissue of the medical subject that has been extracted from the medical subject (e.g., a sample obtained from a breast lesion, or from a colon of a medical subject known or suspected of having colon cancer, or from a liver of a medical subject known or suspected of having liver cancer, or so forth, via a biopsy procedure or other sample extraction procedure). The tissue of which a sample is extracted may also be metastatic tissue, e.g. (suspected) malignant tissue originating from the colon, breast, liver, or other organ that has spread outside of the colon, breast, liver, or other organ. In some cases, the tissue sample may be circulating tumor cells, that is, tumor cells that have entered the bloodstream and may be extracted as the extracted tissue sample using suitable isolation techniques. The extracted sample is processed by the laboratory to generate genomic or proteomic information. For example, the extracted sample may be processed using a microarray (also variously referred to in the art as a gene chip, DNA chip, biochip, or so forth) or by quantitative polymerase chain reaction (qPCR) processing to measure probative genomic or proteomic information such as expression levels of genes of interest, for example in the form of a level of messenger ribonucleic acid (mR A) that is transcribed from the gene, or a level of a protein that is translated from the mRNA transcribed from the gene. As another example, the extracted sample may be processed by a gene sequencing laboratory to generate sequences for deoxyribonucleic acid (DNA), or to generate an RNA sequence, copy number variation, or so forth. Other contemplated measurement approaches include immunohistochemistry (IHC), cytology, fluorescence in situ hybridization (FISH), proximity ligation assay or so forth, performed on a pathology slide. Other information that can be generated by microarray processing, mass
spectrometry, gene sequencing, or other laboratory techniques includes methylation information. Various combinations of such genomic and/or proteomic measurements may also be performed.
In some embodiments, the medical laboratories 18 perform a number of standardized data acquisitions on the extracted sample of the tissue of the medical subject, so as to generate a large quantity of genomic and/or proteomic data. For example, the standardized data acquisition techniques may generate an (optionally aligned) DNA sequence for one or more chromosomes or chromosome portions, or for the entire genome of the tissue. Applying a standard microarray can generate thousands or tens of thousands of data items such as expression levels for a large number of genes, various methylation data, and so forth. This plethora of genomic and/or proteomic data, or selected portions thereof, are input to the CDS system 10 to be processed so as to develop clinically useful information for formulating clinical decision support recommendations.
The disclosed CDS systems and related methods relate to processing of genomic and/or proteomic data to assess activity of various cellular signaling pathways. However, it is to be understood that the disclosed CDS systems (e.g., CDS system 10) may optionally further include diverse additional capabilities, such as generating clinical decision support recommendations in accordance with stored clinical guidelines based on various patient data such as vital sign monitoring data, patient history data, patient demographic data (e.g., gender, age, or so forth), patient medical imaging data, or so forth. Alternatively, in some embodiments the capabilities of the CDS system 10 may be limited to only performing genomic and/or proteomic data analyses to assess cellular signaling pathways as disclosed herein.
With continuing reference to exemplary Figure 20, the CDS system 10 infers activity of a cellular signaling pathway in the tissue of the medical subject based at least on, but not restricted to, expression levels of target genes of the cellular signaling pathway measured in the extracted sample, and determines whether the cellular signaling pathway is operating abnormally in the tissue of the medical subject based on this inferred activity. Examples disclosed herein relate to the Wnt, ER, AR and Hedgehog pathways as illustrative cellular signaling pathways. These pathways are of interest in various areas of oncology because loss of regulation of the pathways can be a cause of proliferation of a cancer. There are about 10-15 relevant signaling pathways, and each cancer is driven by in principle one dominant pathway being deregulated. Without being limited to any particular theory of operation these pathways regulate cell proliferation, and consequentially a loss of regulation of these pathways in cancer cells can lead to the pathway being "always on" thus accelerating the proliferation of cancer cells, which in turn manifests as a growth, invasion or metastasis (spread) of the cancer.
Measurement of mRNA expression levels of genes that encode for regulatory proteins of the cellular signaling pathway, such as an intermediate protein that is part of a protein cascade forming the cellular signaling pathway, is an indirect measure of the regulatory protein expression level and may or may not correlate strongly with the actual regulatory protein expression level (much less with the overall activity of the cellular signaling pathway). The cellular signaling pathway directly regulates the transcription of the target genes - hence, the expression levels of mR A transcribed from the target genes is a direct result of this regulatory activity. Hence, the CDS system 10 infers activity of the cellular signaling pathway (e.g., the Wnt, ER, AR and Hedgehog pathways) based at least on expression levels of target genes (mRNA or protein level as a surrogate measurement) of the cellular signaling pathway. This ensures that the CDS system 10 infers the activity of the pathway based on direct information provided by the measured expression levels of the target genes.
However, although, as disclosed herein, being effective for assessing activity of the overall pathways, the measured expression levels 20 of target genes of the pathways are not especially informative as to why the pathways are operating abnormally (if indeed that is the case). Said another way, the measured expression levels 20 of target genes of a pathway can indicate that the pathway is operating abnormally, but do not indicate what portion of the pathway is malfunctioning (e.g., lacks sufficient regulation) in order to cause the overall pathway to operate abnormally.
Accordingly, if the CDS system 10 detects abnormal activity of a particular pathway, the CDS system 10 then optionally makes use of other information provided by the medical laboratories 18 for the extracted sample, such as aligned genetic sequences 22 and/or measured expression level(s) for one or more regulatory genes of the pathway 24, or select the diagnostic test to be performed next in order to assess what portion of the pathway is malfunctioning. To maximize efficiency, in some embodiments this optional assessment of why the pathway is malfunctioning is performed only if the analysis of the measured expression levels 20 of target genes of the pathway indicates that the pathway is operating abnormally. In other embodiments, this assessment is integrated into the probabilistic analysis of the cellular signaling pathway described herein.
In embodiments in which the CDS system 10 assesses what portion of the pathway is malfunctioning, and is successful in doing so, the additional information enables the CDS system 10 to recommend prescribing a drug targeting for the specific malfunction (recommendation 26 shown in Figure 20). If no specific pathway malfunction is identified (either because the optional additional assessment is not performed or because that assessment fails to identify any particular portion of the pathway that is malfunctioning), then the CDS system 10 can provide a default recommendation 28 recommending the prescription of a general suppression drug for this particular pathway (assuming that the abnormal pathway activity is overly high activity). Example 16: A kit and analysis tools to measure pathway activity
The set of target genes which are found to best indicate specific pathway activity, based on microarray/R A sequencing based investigation using the Bayesian model, can be translated into a multiplex quantitative PCR assay to be performed on a tissue or cell sample. To develop such an FDA-approved test for pathway activity, development of a standardized test kit is required, which needs to be clinically validated in clinical trials to obtain regulatory approval.
In general, it is to be understood that while examples pertaining to the Wnt, the ER, the AR and/or the Hedgehog pathway(s) are provided as illustrative examples, the approaches for cellular signaling pathway analysis disclosed herein are readily applied to other cellular signaling pathways besides these pathways, such as to intracellular signaling pathways with receptors in the cell membrane (cf. above) and intracellular signaling pathways with receptors inside the cell (cf. above). In addition: This application describes several preferred embodiments. Modifications and alterations may occur to others upon reading and understanding the preceding detailed description. It is intended that the application be construed as including all such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Literature:
de Sousa E Melo F, C. S. (2011). Methylation of cancer-stem-cell-associated Wnt target genes predicts poor prognosis in colorectal cancer patients. Cell Stem Cell, 476-485
Hatzis P, v. d. (2008). Genome-wide pattern of TCF7L2/TCF4 chromatin occupancy in colorectal cancer cells. Mol Cell Biol, 2732-2744
Neapolitan, R. (2004). Learning Bayesian networks. Pearson Prentice Hall Nusse, R. (2012, May 1). Wnt target genes. Retrieved from The Wnt homepage:
http://www.stanford.edu/group/nusselab/cgi-bin/wnt/target_genes
Soderberg O, G. M. (2006). Direct observation of individual endogenous protein complexes in situ by proximity ligation. Nat Methods., 995-1000 van de Wetering M, S. E.-P.-F. (2002). The beta-catenin/TCF-4 complex imposes a crypt progenitor phenotype on colorectal cancer cells. Cell, 241-250.

Claims

CLAIMS:
1. A method comprising:
inferring activity of one or more cellular signaling pathway(s) in tissue of a medical subject based at least on the expression level(s) (20) of one or more target gene(s) of the cellular signaling pathway(s) measured in an extracted sample of the tissue of the medical subject, wherein the inferring comprises:
inferring activity of the cellular signaling pathway(s) in the tissue of the medical subject by evaluating at least a portion of a probabilistic model (40-1, 40-7), preferably a Bayesian network (40-1, 40-7), representing the cellular signaling pathway(s) for a set of inputs including at least the expression level(s) (20) of the one or more target genes of the cellular signaling pathway(s) measured in the extracted sample of the tissue of the medical subject;
estimating a level (46) in the tissue of the medical subject of at least one transcription factor (TF) element, the at least one TF element controlling transcription of the one or more target gene(s) of the cellular signaling pathway(s), the estimating being based at least in part on conditional probabilities relating the at least one TF element and the expression level(s) (20) of the one or more target gene(s) of the cellular signaling pathway(s) measured in the extracted sample of the tissue of the medical subject;
inferring activity of the cellular signaling pathway(s) based on the estimated level in the tissue sample of the transcription factor; and
determining whether the cellular signaling pathway(s) is/are operating abnormally in the tissue of the medical subject based on the inferred activity of the cellular signaling pathway(s) in the tissue of the medical subject;
wherein the inferring is performed by a digital processing device (12) using the probabilistic model (40-1, 40-7) of the cellular signaling pathway(s).
2. A method according claim 1, wherein the inferring comprises:
estimating a level (46) in the tissue of the medical subject of at least one transcription factor (TF) element represented by a TF node of the probabilistic model, the TF element controlling transcription of the one or more target gene(s) of the cellular signaling pathway(s), the estimating being based at least in part on conditional probabilities of the probabilistic model (40-1, 40-7) relating the TF node and nodes in the probabilistic model representing the one or more target gene(s) of the cellular signaling pathway(s) measured in the extracted sample of the tissue of the medical subject,
and wherein the inferring is preferably performed by using a Bayesian network (40-1, 40-7) comprising nodes representing information about the signaling pathway(s) and conditional probability relationships between connected nodes of the Bayesian network.
3. The method of any one of claim 1 or 2, wherein the cellular signaling pathway(s) comprise(s) a Wnt pathway, an ER pathway, an AR pathway and/or a
Hedgehog pathway.
4. The method of claim 3, wherein the inferring comprises:
inferring activity of a Wnt pathway in the tissue of the medical subject based at least on expression levels (20) of one or more, preferably at least three, target gene(s) of the Wnt pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising: KIAA1199, AXIN2, RNF43, TBX3, TDGFl, SOX9, ASCL2, IL8, SP5, ZNRF3, KLF6, CCND1, DEFA6 and FZD7.
5. The method of claim 3 or 4, wherein the inferring (also) comprises:
inferring activity of an ER pathway in the tissue of the medical subject based at least on expression levels (20) of one or more, preferably at least three, target gene(s) of the ER pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising: CDH26, SGK3, PGR, GREBl, CA12, XBP1, CELSR2, WISP2, DSCAM, ERBB2, CTSD, TFF1 and NRIPl .
6. The method of any of claims 3 to 5, wherein the inferring (also) comprises: inferring activity of a Hedgehog pathway in the tissue of the medical subject based at least on expression levels (20) of one or more, preferably at least three, target gene(s) of the Hedgehog pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising: GLI1 , PTCH1 , PTCH2, IGFBP6, SPP1 ,
CCND2, FST, FOXLl, CFLAR, TSC22D1, RAB34, S100A9, S100A7, MYCN, FOXMl, GLI3, TCEA2, FYN and CTSL1.
7. The method of any of claims 3 to 6, wherein the inferring (also) comprises: inferring activity of an AR pathway in the tissue of the medical subject based at least on expression levels (20) of one or more, preferably at least three, target gene(s) of the AR pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising: KLK2, PMEPA1, TMPRSS2, NKX3 1, ABCC4, KLK3, FKBP5, ELL2, UGT2B15, DHCR24, PPAP2A, NDRG1, LRIG1, CREB3L4, LCP1, GUCY1A3, AR and EAF2.
8. The method of claim 4, wherein the inferring is further based on expression levels (20) of at least one target gene of the Wnt pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising: NKD1, OAT, FATl, LEFl, GLUL, REGIB, TCF7L2, COL18A1, BMP7, SLC1A2, ADRA2C, PPARG, DK 1, HNFIA and LECT2.
9. The method of claim 5, wherein the inferring is further based on expression levels (20) of at least one target gene of the ER pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising: AP1B1, ATP5J, COL18A1, COX7A2L, EBAG9, ESR1, HSPB1, IGFBP4, KRT19, MYC, NDUFV3, PISD, PREDM15, PTMA, RARA, SOD1 and TRIM25.
10. The method of claim 6, wherein the inferring is further based on expression levels (20) of at least one target gene of the Hedgehog pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising: BCL2, FOXA2, FOXF1, G19, HHIP, IL1R2, JAG2, JUP, MIF, MYLK, NKX2.2, NKX2.8, PITRM1 and TOM1.
11. The method of claim 7, wherein the inferring is further based on expression levels (20) of at least one target gene of the AR pathway measured in the extracted sample of the tissue of the medical subject selected from the group comprising: APP, NTS, PLAU, CDKN1A, DRG1, FGF8, IGF1, PRKACB, PTPN1, SGK1 and TACC2.
12. The method of any one of claims 1-11, further comprising:
recommending prescribing a drug for the medical subject that corrects for abnormal operation of the cellular signaling pathway(s);
wherein the recommending is performed only if the cellular signaling pathway(s) is/are determined to be operating abnormally in the tissue of the medical subject based on the inferred activity of the cellular signaling pathway(s).
13. The method of any of claims 1-11, wherein the method is used in at least one of the following activities:
diagnosis based on the inferred activity of the cellular signaling pathways(s) in the tissue of the medical subject;
prognosis based on the inferred activity of the cellular signaling pathways(s) in the tissue of the medical subject;
drug prescription based on the inferred activity of the cellular signaling pathways(s) in the tissue of the medical subject;
prediction of drug efficacy based on the inferred activity of the cellular signaling pathways(s) in the tissue of the medical subject;
prediction of adverse effects based on the inferred activity of the cellular signaling pathways(s) in the tissue of the medical subject;
monitoring of drug efficacy;
drug development;
assay development;
pathway research;
cancer staging;
enrollment of the medical subject in a clinical trial based on the inferred activity of the cellular signaling pathways(s) in the tissue of the medical subject;
selection of subsequent test to be performed, and
selection of companion diagnostics tests.
14. The method of any one of claims 1-13 comprising:
inferring activity of a Wnt pathway in tissue of a medical subject based at least on expression levels (20) of two, three or more target genes of a set of target genes of the Wnt pathway measured in an extracted sample of the tissue of the medical subject
and/or
inferring activity of an ER pathway in tissue of a medical subject based at least on expression levels (20) of two, three or more target genes of a set of target genes of the ER pathway measured in an extracted sample of the tissue of the medical subject
and/or
inferring activity of a Hedgehog pathway in tissue of a medical subject based at least on expression levels (20) of two, three or more target genes of a set of target genes of the Hedgehog pathway measured in an extracted sample of the tissue of the medical subject,
and/or
inferring activity of an AR pathway in tissue of a medical subject based at least on expression levels (20) of two, three or more target genes of a set of target genes of the AR pathway measured in an extracted sample of the tissue of the medical subject.
15. The method of claim 14, wherein
the set of target genes of the Wnt pathway includes at least nine, preferably all target genes selected from the group comprising: KIAA1199, AXIN2, RNF43, TBX3, TDGF1, SOX9, ASCL2, IL8, SP5, ZNRF3, KLF6, CCND1, DEFA6 and FZD7,
and/or
the set of target genes of the ER pathway includes at least nine, preferably all target genes selected from the group comprising: CDH26, SGK3, PGR, GREB1, CA12, XBP1, CELSR2, WISP2, DSC AM, ERBB2, CTSD, TFF1 and NRIP1,
and/or
the set of target genes of the Hedgehog pathway includes at least nine, preferably all target genes selected from the group comprising: GLI1, PTCH1, PTCH2, IGFBP6, SPP1, CCND2, FST, FOXL1, CFLAR, TSC22D1, RAB34, S100A9, S100A7, MYCN, FOXM1, GLI3, TCEA2, FYN and CTSL1,
and/or the set of target genes of the AR pathway includes at least nine, preferably all target genes selected from the group comprising: KLK2, PMEPA1, TMPRSS2, NKX3 1, ABCC4, KLK3, FKBP5, ELL2, UGT2B15, DHCR24, PPAP2A, NDRG1, LRIG1, CREB3L4, LCP1, GUCY1A3, AR and EAF2.
16. The method of claim 15, wherein
the set of target genes of the Wnt pathway further includes at least one target gene selected from the group comprising: NKD1, OAT, FAT1, LEF1, GLUL, REGIB,
TCF7L2, COL18A1, BMP7, SLC1A2, ADRA2C, PPARG, DKKl, HNFIA, and LECT2, and/or
the set of target genes of the ER pathway further includes at least one target gene selected from the group comprising: AP1B1, ATP5J, COL18A1, COX7A2L, EBAG9, ESR1, HSPB1, IGFBP4, KRT19, MYC, NDUFV3, PISD, PREDM15, PTMA, RARA, SOD1 and TRIM25,
and/or
the set of target genes of the Hedgehog pathway further includes at least one target gene selected from the group comprising: BCL2, FOXA2, FOXF1, G19, HHIP, IL1R2, JAG2, JUP, MIF, MYLK, NKX2.2, NKX2.8, PITRM1 and TOM1,
and/or
the set of target genes of the AR pathway further includes at least one target gene selected from the group comprising: APP, NTS, PLAU, CDKNIA, DRGl , FGF8, IGFl , PRKACB, PTPN1, SGK1 and TACC2.
17. An apparatus comprising a digital processor (12) configured to perform a method as set forth in any one of claims 1-16.
18. A non-transitory storage medium storing instructions that are executable by a digital processing device (12) to perform a method as set forth in any one of claims 1-16.
19. A computer program comprising program code means for causing a digital processing device (12) to perform a method as set forth in any one of claims 1-16.
20. Use of a non-transitory storage medium as set forth in claim 18 or a computer program as set forth in claim 19 for specific diagnosis of a disease or predicting response to a drug in a specific cancer type at a specific stage.
EP12751371.1A 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression Active EP2734643B1 (en)

Priority Applications (6)

Application Number Priority Date Filing Date Title
EP20209212.8A EP3831957B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP12751371.1A EP2734643B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP18180343.8A EP3418396B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
DK16192584.7T DK3173489T3 (en) 2011-07-19 2012-07-19 ASSESSMENT OF THE ACTIVITY OF A CELLULAR SIGNAL ROUTE USING PROBALISTIC MODELING OF TARGET EXPRESSION
EP22177533.1A EP4130291B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP16192584.7A EP3173489B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US201161509137P 2011-07-19 2011-07-19
EP11178148A EP2549399A1 (en) 2011-07-19 2011-08-19 Assessment of Wnt pathway activity using probabilistic modeling of target gene expression
EP12751371.1A EP2734643B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
PCT/IB2012/053686 WO2013011479A2 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression

Related Child Applications (5)

Application Number Title Priority Date Filing Date
EP20209212.8A Division EP3831957B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP18180343.8A Division EP3418396B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP22177533.1A Division EP4130291B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP16192584.7A Division-Into EP3173489B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP16192584.7A Division EP3173489B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression

Publications (2)

Publication Number Publication Date
EP2734643A2 true EP2734643A2 (en) 2014-05-28
EP2734643B1 EP2734643B1 (en) 2016-12-07

Family

ID=44582424

Family Applications (6)

Application Number Title Priority Date Filing Date
EP11178148A Ceased EP2549399A1 (en) 2011-07-19 2011-08-19 Assessment of Wnt pathway activity using probabilistic modeling of target gene expression
EP16192584.7A Active EP3173489B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP20209212.8A Active EP3831957B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP22177533.1A Active EP4130291B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP18180343.8A Active EP3418396B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP12751371.1A Active EP2734643B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression

Family Applications Before (5)

Application Number Title Priority Date Filing Date
EP11178148A Ceased EP2549399A1 (en) 2011-07-19 2011-08-19 Assessment of Wnt pathway activity using probabilistic modeling of target gene expression
EP16192584.7A Active EP3173489B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP20209212.8A Active EP3831957B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP22177533.1A Active EP4130291B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
EP18180343.8A Active EP3418396B1 (en) 2011-07-19 2012-07-19 Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression

Country Status (12)

Country Link
US (2) US11443831B2 (en)
EP (6) EP2549399A1 (en)
JP (3) JP6204354B2 (en)
KR (1) KR102064004B1 (en)
CN (2) CN109852671A (en)
BR (1) BR112014000965B8 (en)
DK (3) DK2734643T3 (en)
ES (3) ES2874452T3 (en)
IN (1) IN2014CN00675A (en)
MX (1) MX352827B (en)
RU (1) RU2719194C2 (en)
WO (1) WO2013011479A2 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113278700A (en) * 2021-06-04 2021-08-20 浙江省肿瘤医院 Primer group and kit for breast cancer typing and prognosis prediction
CN116083590A (en) * 2023-03-23 2023-05-09 广州爱仁生物医药科技有限公司 Gene detection kit and preparation method and application thereof

Families Citing this family (71)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11660317B2 (en) 2004-11-08 2023-05-30 The Johns Hopkins University Compositions comprising cardiosphere-derived cells for use in cell therapy
EP2549399A1 (en) 2011-07-19 2013-01-23 Koninklijke Philips Electronics N.V. Assessment of Wnt pathway activity using probabilistic modeling of target gene expression
KR20140132712A (en) 2012-02-28 2014-11-18 노파르티스 아게 Cancer patient selection for administration of wnt signaling inhibitors using rnf43 mutation status
EP3563859B1 (en) 2012-08-13 2021-10-13 Cedars-Sinai Medical Center Cardiosphere-derived exosomes for tissue regeneration
CA2896414C (en) * 2012-12-26 2023-06-20 Koninklijke Philips N.V. Assessment of cellular signaling pathway activity using linear combination(s) of target gene expressions
TWI582239B (en) * 2013-03-11 2017-05-11 諾華公司 Markers associated with wnt inhibitors
US11309059B2 (en) 2013-04-26 2022-04-19 Koninklijke Philips N.V. Medical prognosis and prediction of treatment response using multiple cellular signalling pathway activities
AU2014302070B2 (en) * 2013-06-28 2016-09-15 Nantomics, Llc Pathway analysis for identification of diagnostic tests
US20160237163A1 (en) * 2013-10-02 2016-08-18 Suri Technologies Ltd. Patient-specific immunotherapy for treating heterogeneous tumors
US11261495B2 (en) 2014-01-03 2022-03-01 Koninklijke Philips N.V. Assessment of the PI3K cellular signaling pathway activity using mathematical modelling of target gene expression
US9810762B2 (en) * 2014-09-11 2017-11-07 Google Inc. Calculating mean wireless signal strengths using a gaussian process approach incorporating predictive standard deviations
US9838847B2 (en) 2014-09-11 2017-12-05 Google LLP Data driven evaluation and rejection of trained Gaussian process-based wireless mean and standard deviation models
US9880257B2 (en) * 2014-09-11 2018-01-30 Google Llc Gaussian process-based approach for identifying correlation between wireless signals
US10557175B2 (en) 2014-09-19 2020-02-11 The Provost, Fellows, Scholars And Other Members Of Board Of Trinity College Dublin Method of predicting risk of recurrence of cancer
JP6878274B2 (en) 2014-10-03 2021-05-26 シーダーズ−サイナイ・メディカル・センターＣｅｄａｒｓ−Ｓｉｎａｉ Ｍｅｄｉｃａｌ Ｃｅｎｔｅｒ Myocardial cell-derived cells and exosomes secreted by such cells in the treatment of muscular dystrophy
ES2838923T3 (en) 2014-10-24 2021-07-02 Koninklijke Philips Nv Medical prognosis and prediction of response to treatment using multiple activities of the cell signaling pathway
DK3210142T3 (en) * 2014-10-24 2020-11-16 Koninklijke Philips Nv ASSESSMENT OF TGF CELLULAR SIGNALING VEHICLE ACTIVITY USING MATHEMATICAL MODELING OF MOAL EXPRESSION
CN108064311B (en) * 2014-10-24 2022-10-28 皇家飞利浦有限公司 Medical prognosis and prediction of treatment response using multiple cell signaling pathway activities
CN105760707B (en) * 2014-12-18 2018-07-31 中国科学院大连化学物理研究所 Cytogene translation process modeling method
WO2017029215A1 (en) * 2015-08-14 2017-02-23 Koninklijke Philips N.V. Assessment of nfkb cellular signaling pathway activity using mathematical modelling of target gene expression
EP3341875A1 (en) * 2015-08-27 2018-07-04 Koninklijke Philips N.V. An integrated method and system for identifying functional patient-specific somatic aberations using multi-omic cancer profiles
CN106897578B (en) * 2015-12-15 2020-02-14 中国科学院大连化学物理研究所 Modeling method for cell gene translation process
WO2017123662A1 (en) 2016-01-11 2017-07-20 Cedars-Sinai Medical Center Cardiosphere-derived cells and exosomes secreted by such cells in the treatment of heart failure with preserved ejection fraction
US20170329914A1 (en) * 2016-05-11 2017-11-16 International Business Machines Corporation Predicting Personalized Cancer Metastasis Routes, Biological Mediators of Metastasis and Metastasis Blocking Therapies
WO2017210652A1 (en) 2016-06-03 2017-12-07 Cedars-Sinai Medical Center Cdc-derived exosomes for treatment of ventricular tachyarrythmias
US10621236B2 (en) * 2016-09-16 2020-04-14 At&T Intellectual Property I, L.P. Concept based querying of graph databases
US11541078B2 (en) 2016-09-20 2023-01-03 Cedars-Sinai Medical Center Cardiosphere-derived cells and their extracellular vesicles to retard or reverse aging and age-related disorders
EP3538673A4 (en) 2016-11-11 2019-12-04 University of Pittsburgh- Of the Commonwealth System of Higher Education Identification of instance-specific somatic genome alterations with functional impact
CN110382521B (en) 2016-11-25 2024-07-05 皇家飞利浦有限公司 Method for differentiating tumor-inhibiting FOXO activity from oxidative stress
CN108300783A (en) * 2017-01-11 2018-07-20 上海易毕恩基因科技有限公司 The method of gene marker for screening intestinal cancer and/or gastric cancer, the gene marker and application thereof screened with this method
CN106755464A (en) * 2017-01-11 2017-05-31 上海易毕恩基因科技有限公司 For the method for screening the gene marker of intestinal cancer and/or stomach cancer, the gene marker and application thereof that is screened with the method
US11759482B2 (en) 2017-04-19 2023-09-19 Cedars-Sinai Medical Center Methods and compositions for treating skeletal muscular dystrophy
CN107184991B (en) * 2017-05-03 2020-05-26 武汉爱博泰克生物科技有限公司 Application of PRDM15 gene sequence in preparation of NF-kB inhibitor
CN107164508A (en) * 2017-06-16 2017-09-15 上海易毕恩基因科技有限公司 Gene marker for detecting liver cancer and application thereof
EP3431582A1 (en) 2017-07-18 2019-01-23 Koninklijke Philips N.V. Cell culturing materials
CN107201411A (en) * 2017-07-27 2017-09-26 上海市长宁区妇幼保健院 MYLK genes as diagnosis of endometrial carcinoma mark
CN107385051A (en) * 2017-08-04 2017-11-24 上海易毕恩生物技术有限公司 For detecting liver tumour good pernicious gene marker, kit and detection method
EP3462348A1 (en) 2017-09-28 2019-04-03 Koninklijke Philips N.V. Bayesian inference
CN111587293A (en) 2017-10-02 2020-08-25 皇家飞利浦有限公司 Determining immune cell type and functional status of immune response
EP3462349A1 (en) * 2017-10-02 2019-04-03 Koninklijke Philips N.V. Assessment of notch cellular signaling pathway activity using mathematical modelling of target gene expression
EP3461915A1 (en) 2017-10-02 2019-04-03 Koninklijke Philips N.V. Assessment of jak-stat1/2 cellular signaling pathway activity using mathematical modelling of target gene expression
EP3461916A1 (en) 2017-10-02 2019-04-03 Koninklijke Philips N.V. Assessment of jak-stat3 cellular signaling pathway activity using mathematical modelling of target gene expression
EP3502279A1 (en) 2017-12-20 2019-06-26 Koninklijke Philips N.V. Assessment of mapk-ap 1 cellular signaling pathway activity using mathematical modelling of target gene expression
US11660355B2 (en) 2017-12-20 2023-05-30 Cedars-Sinai Medical Center Engineered extracellular vesicles for enhanced tissue delivery
US11475995B2 (en) * 2018-05-07 2022-10-18 Perthera, Inc. Integration of multi-omic data into a single scoring model for input into a treatment recommendation ranking
CN108875009B (en) * 2018-06-16 2021-10-22 上海豹云网络信息服务有限公司 Method and system for dynamically processing data file
US11354591B2 (en) 2018-10-11 2022-06-07 International Business Machines Corporation Identifying gene signatures and corresponding biological pathways based on an automatically curated genomic database
WO2020124585A1 (en) * 2018-12-21 2020-06-25 北京哲源科技有限责任公司 Method for acquiring intracellular deterministic event, electronic device, and storage medium
EP3812474A1 (en) 2019-10-22 2021-04-28 Koninklijke Philips N.V. Methods of prognosis in high-grade serous ovarian cancer
CN113785075A (en) 2019-05-03 2021-12-10 皇家飞利浦有限公司 Method for prognosis of high-grade serous ovarian cancer
EP3965800A4 (en) * 2019-05-08 2023-02-15 Cedars-Sinai Medical Center Therapeutically active cells and exosomes
EP3739588A1 (en) 2019-05-13 2020-11-18 Koninklijke Philips N.V. Assessment of multiple signaling pathway activity score in airway epithelial cells to predict airway epithelial abnormality and airway cancer risk
AU2020334901A1 (en) * 2019-08-16 2022-03-24 Tempus Ai, Inc. Systems and methods for detecting cellular pathway dysregulation in cancer specimens
EP3822368A1 (en) 2019-11-14 2021-05-19 Koninklijke Philips N.V. Assessment of pr cellular signaling pathway activity using mathematical modelling of target gene expression
EP3835433A1 (en) 2019-12-12 2021-06-16 Koninklijke Philips N.V. Notch signaling pathway activity as prognostic marker in bladder cancer
AU2020403276A1 (en) 2019-12-12 2022-06-09 Tempus Ai, Inc. Real-world evidence of diagnostic testing and treatment patterns in U.S. breast cancer patients
CN111139300B (en) * 2020-02-19 2021-08-17 伯克利南京医学研究有限责任公司 Application of group of colon cancer prognosis related genes
EP3882363A1 (en) 2020-03-17 2021-09-22 Koninklijke Philips N.V. Prognostic pathways for high risk sepsis patients
EP3978628A1 (en) 2020-10-01 2022-04-06 Koninklijke Philips N.V. Prognostic pathways for viral infections
EP4136261A1 (en) 2020-04-16 2023-02-22 InnoSIGN B.V. Prognostic pathways for viral infections
CN111739586B (en) * 2020-06-17 2024-04-05 浙江大学 Model for predicting cell proliferation activity by using 87 genes as biomarkers
EP3940704A1 (en) 2020-07-14 2022-01-19 Koninklijke Philips N.V. Method for determining the differentiation state of a stem cell
EP3960875A1 (en) 2020-08-28 2022-03-02 Koninklijke Philips N.V. Pcr method and kit for determining pathway activity
EP3965119A1 (en) 2020-09-04 2022-03-09 Koninklijke Philips N.V. Methods for estimating heterogeneity of a tumour based on values for two or more genome mutation and/or gene expression related parameter, as well as corresponding devices
EP3974540A1 (en) 2020-09-25 2022-03-30 Koninklijke Philips N.V. Method for predicting immunotherapy resistance
EP4015651A1 (en) 2020-12-17 2022-06-22 Koninklijke Philips N.V. Treatment prediction and effectiveness of anti-tnf alpha treatment in ibd patients
EP4039825A1 (en) 2021-02-09 2022-08-10 Koninklijke Philips N.V. Comparison and standardization of cell and tissue culture
WO2022189530A1 (en) 2021-03-11 2022-09-15 Koninklijke Philips N.V. Prognostic pathways for high risk sepsis patients
WO2023084039A1 (en) 2021-11-15 2023-05-19 Innosign B.V. Assessment of gr cellular signaling pathway activity using mathematical modelling of target gene expression
CN113963745A (en) * 2021-12-07 2022-01-21 国际竹藤中心 Method for constructing plant development molecule regulation network and application thereof
WO2024033063A1 (en) 2022-08-12 2024-02-15 Innosign B.V. Prediction and monitoring of immunotherapy in cancer

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6566130B1 (en) * 2000-01-28 2003-05-20 Henry M. Jackson Foundation For The Advancement Of Military Medicine Androgen-regulated gene expressed in prostate tissue
US7371736B2 (en) 2001-11-07 2008-05-13 The Board Of Trustees Of The University Of Arkansas Gene expression profiling based identification of DKK1 as a potential therapeutic targets for controlling bone loss
EP1380644A1 (en) 2002-07-08 2004-01-14 Kylix B.V. The use of specified TCF target genes to identify drugs for the treatment of cancer, in particular colorectal cancer, in which TCF/beta-catenin/WNT signalling plays a central role
AU2003260973A1 (en) 2002-09-05 2004-03-29 Toolgen, Inc. Bioinformatics analysis of cellular effects of artificial transcription factors
WO2006079092A2 (en) * 2005-01-24 2006-07-27 The Board Of Trustees Of The Leland Stanford Junior University Method for modeling cell signaling systems by means of bayesian networks
WO2006124836A1 (en) 2005-05-13 2006-11-23 Duke University Gene expression signatures for oncogenic pathway deregulation
WO2007056344A2 (en) 2005-11-07 2007-05-18 Scanscout, Inc. Techiques for model optimization for statistical pattern recognition
EP2027549A2 (en) 2006-05-17 2009-02-25 The Trustees Of Boston University Method to determine transcriptional regulation pathways in organisms
WO2008086182A2 (en) * 2007-01-04 2008-07-17 University Of Rochester Use of gene signatures to design novel cancer treatment regimens
JP2010178650A (en) 2009-02-04 2010-08-19 Univ Of Tokyo Test method for predicting recurrence of solid cancer and recurrence prophylactic
US20120252689A1 (en) 2011-04-01 2012-10-04 Zhong Wu Gene expression signature for wnt/b-catenin signaling pathway and use thereof
AU2012250498A1 (en) * 2011-05-05 2013-11-21 Clinical Genomics Pty. Ltd. A method of diagnosing neoplasms
EP2549399A1 (en) 2011-07-19 2013-01-23 Koninklijke Philips Electronics N.V. Assessment of Wnt pathway activity using probabilistic modeling of target gene expression

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ROGERS SIMON ET AL: "Bayesian model-based inference of transcription factor activity", BMC BIOINFORMATICS, BIOMED CENTRAL, LONDON, GB, vol. 8, no. Suppl 2, 3 May 2007 (2007-05-03), page S2, XP021021916, ISSN: 1471-2105, DOI: 10.1186/1471-2105-8-S2-S2 *
See also references of WO2013011479A2 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113278700A (en) * 2021-06-04 2021-08-20 浙江省肿瘤医院 Primer group and kit for breast cancer typing and prognosis prediction
CN116083590A (en) * 2023-03-23 2023-05-09 广州爱仁生物医药科技有限公司 Gene detection kit and preparation method and application thereof
CN116083590B (en) * 2023-03-23 2024-03-22 雄安妙心医学检验有限公司 Gene detection kit and preparation method and application thereof

Also Published As

Publication number Publication date
BR112014000965A2 (en) 2017-06-13
KR20140046464A (en) 2014-04-18
EP3831957A1 (en) 2021-06-09
CN103649337A (en) 2014-03-19
RU2719194C2 (en) 2020-04-17
JP2014520566A (en) 2014-08-25
DK2734643T3 (en) 2017-03-20
JP6204354B2 (en) 2017-09-27
ES2617195T3 (en) 2017-06-15
JP6807430B2 (en) 2021-01-06
EP2734643B1 (en) 2016-12-07
MX352827B (en) 2017-12-11
EP4130291C0 (en) 2024-01-10
EP4130291A1 (en) 2023-02-08
US11443831B2 (en) 2022-09-13
JP6783729B2 (en) 2020-11-11
EP3173489A1 (en) 2017-05-31
BR112014000965B1 (en) 2021-08-24
US20230105263A1 (en) 2023-04-06
ES2700617T3 (en) 2019-02-18
EP2549399A1 (en) 2013-01-23
CN103649337B (en) 2018-11-13
KR102064004B1 (en) 2020-01-08
US20140156200A1 (en) 2014-06-05
EP3173489B1 (en) 2018-09-19
IN2014CN00675A (en) 2015-04-03
CN109852671A (en) 2019-06-07
RU2014106026A (en) 2015-08-27
ES2874452T3 (en) 2021-11-05
JP2019129850A (en) 2019-08-08
WO2013011479A3 (en) 2013-03-14
BR112014000965B8 (en) 2022-11-08
EP3418396A1 (en) 2018-12-26
EP3831957B1 (en) 2022-09-07
DK3173489T3 (en) 2018-12-10
MX2014000557A (en) 2014-05-01
EP4130291B1 (en) 2024-01-10
DK3418396T3 (en) 2021-05-25
WO2013011479A2 (en) 2013-01-24
JP2017205129A (en) 2017-11-24
EP3418396B1 (en) 2021-03-24

Similar Documents

Publication Publication Date Title
US20230105263A1 (en) Assessment of cellular signaling pathway activity using probabilistic modeling of target gene expression
AU2020202706B2 (en) Assessment of cellular signaling pathway activity using linear combination(s) of target gene expressions
AU2014259459B2 (en) Medical prognosis and prediction of treatment response using multiple cellular signalling pathway activities

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20140219

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

17Q First examination report despatched

Effective date: 20140526

DAX Request for extension of the european patent (deleted)
GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

INTG Intention to grant announced

Effective date: 20160608

GRAJ Information related to disapproval of communication of intention to grant by the applicant or resumption of examination proceedings by the epo deleted

Free format text: ORIGINAL CODE: EPIDOSDIGR1

GRAR Information related to intention to grant a patent recorded

Free format text: ORIGINAL CODE: EPIDOSNIGR71

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE PATENT HAS BEEN GRANTED

INTC Intention to grant announced (deleted)
AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

INTG Intention to grant announced

Effective date: 20161028

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

Ref country code: AT

Ref legal event code: REF

Ref document number: 851776

Country of ref document: AT

Kind code of ref document: T

Effective date: 20161215

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 602012026298

Country of ref document: DE

REG Reference to a national code

Ref country code: CH

Ref legal event code: NV

Representative=s name: FELBER AND PARTNER AG, CH

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: LV

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

REG Reference to a national code

Ref country code: NL

Ref legal event code: FP

REG Reference to a national code

Ref country code: DK

Ref legal event code: T3

Effective date: 20170315

REG Reference to a national code

Ref country code: SE

Ref legal event code: TRGR

REG Reference to a national code

Ref country code: LT

Ref legal event code: MG4D

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: NO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20170307

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20170308

Ref country code: LT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

REG Reference to a national code

Ref country code: AT

Ref legal event code: MK05

Ref document number: 851776

Country of ref document: AT

Kind code of ref document: T

Effective date: 20161207

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: FI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

Ref country code: HR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

Ref country code: RS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

REG Reference to a national code

Ref country code: ES

Ref legal event code: FG2A

Ref document number: 2617195

Country of ref document: ES

Kind code of ref document: T3

Effective date: 20170615

REG Reference to a national code

Ref country code: FR

Ref legal event code: PLFP

Year of fee payment: 6

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: EE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

Ref country code: CZ

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

Ref country code: IS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20170407

Ref country code: SK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

Ref country code: RO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SM

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

Ref country code: PL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

Ref country code: AT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

Ref country code: BG

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20170307

Ref country code: PT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20170407

REG Reference to a national code

Ref country code: DE

Ref legal event code: R097

Ref document number: 602012026298

Country of ref document: DE

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed

Effective date: 20170908

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: LU

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20170719

REG Reference to a national code

Ref country code: FR

Ref legal event code: PLFP

Year of fee payment: 7

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MT

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20170719

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: HU

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT; INVALID AB INITIO

Effective date: 20120719

Ref country code: MC

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CY

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20161207

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: TR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: AL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20161207

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: IT

Payment date: 20210722

Year of fee payment: 10

Ref country code: IE

Payment date: 20210721

Year of fee payment: 10

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: ES

Payment date: 20210809

Year of fee payment: 10

Ref country code: DK

Payment date: 20210722

Year of fee payment: 10

Ref country code: BE

Payment date: 20210726

Year of fee payment: 10

Ref country code: CH

Payment date: 20210722

Year of fee payment: 10

Ref country code: SE

Payment date: 20210722

Year of fee payment: 10

REG Reference to a national code

Ref country code: NL

Ref legal event code: QB

Free format text: DETAILS LICENCE OR PLEDGE: LICENCE, NEW LICENCE REGISTRATION

Name of requester: INNOSIGN B.V.

Effective date: 20220518

Ref country code: NL

Ref legal event code: PD

Owner name: INNOSIGN B.V.; NL

Free format text: DETAILS ASSIGNMENT: CHANGE OF OWNER(S), ASSIGNMENT; FORMER OWNER NAME: KONINKLIJKE PHILIPS N.V.

Effective date: 20220518

REG Reference to a national code

Ref country code: GB

Ref legal event code: 732E

Free format text: REGISTERED BETWEEN 20220714 AND 20220720

REG Reference to a national code

Ref country code: GB

Ref legal event code: 732E

Free format text: REGISTERED BETWEEN 20220721 AND 20220727

REG Reference to a national code

Ref country code: DE

Ref legal event code: R081

Ref document number: 602012026298

Country of ref document: DE

Owner name: INNOSIGN B.V., NL

Free format text: FORMER OWNER: KONINKLIJKE PHILIPS N.V., EINDHOVEN, NL

Ref country code: DE

Ref legal event code: R082

Ref document number: 602012026298

Country of ref document: DE

REG Reference to a national code

Ref country code: DK

Ref legal event code: EBP

Effective date: 20220731

REG Reference to a national code

Ref country code: CH

Ref legal event code: PL

Ref country code: SE

Ref legal event code: EUG

REG Reference to a national code

Ref country code: BE

Ref legal event code: MM

Effective date: 20220731

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20220720

Ref country code: LI

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20220731

Ref country code: CH

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20220731

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: BE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20220731

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IT

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20220719

Ref country code: IE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20220719

Ref country code: DK

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20220731

REG Reference to a national code

Ref country code: ES

Ref legal event code: FD2A

Effective date: 20230905

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: ES

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20220720

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: NL

Payment date: 20240719

Year of fee payment: 13

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: DE

Payment date: 20240719

Year of fee payment: 13

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 20240725

Year of fee payment: 13

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: FR

Payment date: 20240730

Year of fee payment: 13